Econometrics

Econometrics c Michael Creel Version 0.80, February, 2006 D EPT. OF E CONOMICS AND E CONOMIC H ISTORY, U NIVERSITA...

Author: Creel M.

43 downloads 1666 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Econometrics

c

Michael Creel

Version 0.80, February, 2006

D EPT.

OF

E CONOMICS AND E CONOMIC H ISTORY, U NIVERSITAT AUTÒNOMA DE BARCELONA ,

MICHAEL . CREEL @ UAB . ES , H T T P :// P A R E T O . U A B . E S / M C R E E L

Contents List of Figures

14

List of Tables

17

Chapter 1.

18

About this document

1.1. License

19

1.2. Obtaining the materials

19

1.3. An easy way to use LYX and Octave today

20

1.4. Known Bugs

22

Chapter 2.

Introduction: Economic and econometric models

23

Chapter 3.

Ordinary Least Squares

26

3.1. The Linear Model

26

3.2. Estimation by least squares

27

3.3. Geometric interpretation of least squares estimation

30

3.3.1. In X ,Y Space

30

3.3.2. In Observation Space

30

3.3.3. Projection Matrices

32

3.4. Influential observations and outliers

33

3.5. Goodness of fit

35

3.6. The classical linear regression model

38

3.7. Small sample statistical properties of the least squares estimator

40

3.7.1. Unbiasedness

40 3

CONTENTS

4

3.7.2. Normality

41

3.7.3. The variance of the OLS estimator and the Gauss-Markov theorem

42

3.8. Example: The Nerlove model

47

3.8.1. Theoretical background

47

3.8.2. Cobb-Douglas functional form

48

3.8.3. The Nerlove data and OLS

49

Exercises

53

Chapter 4.

Maximum likelihood estimation

54

4.1. The likelihood function

54

4.1.1. Example: Bernoulli trial

56

4.2. Consistency of MLE

58

4.3. The score function

60

4.4. Asymptotic normality of MLE

62

4.6. The information matrix equality

66

4.7. The Cramér-Rao lower bound

68

Exercises

71

Chapter 5.

Asymptotic properties of the least squares estimator

73

5.1. Consistency

73

5.2. Asymptotic normality

74

5.3. Asymptotic efficiency

75

Chapter 6.

Restrictions and hypothesis tests

77

6.1. Exact linear restrictions

77

6.1.1. Imposition

78

6.1.2. Properties of the restricted estimator

82

6.2. Testing

83

CONTENTS

5

6.2.1. t-test

83

6.2.2. F test

86

6.2.3. Wald-type tests

87

6.2.4. Score-type tests (Rao tests, Lagrange multiplier tests)

88

6.2.5. Likelihood ratio-type tests

90

6.3. The asymptotic equivalence of the LR, Wald and score tests

92

6.4. Interpretation of test statistics

96

6.5. Confidence intervals

96

6.6. Bootstrapping

97

6.7. Testing nonlinear restrictions, and the Delta Method

100

6.8. Example: the Nerlove data

104

Chapter 7.

Generalized least squares

111

7.1. Effects of nonspherical disturbances on the OLS estimator

112

7.2. The GLS estimator

113

7.3. Feasible GLS

116

7.4. Heteroscedasticity

118

7.4.1. OLS with heteroscedastic consistent varcov estimation

119

7.4.2. Detection

119

7.4.3. Correction

122

7.4.4. Example: the Nerlove model (again!)

125

7.5. Autocorrelation

130

7.5.1. Causes

130

7.5.2. Effects on the OLS estimator

132

7.5.3. AR(1)

132

7.5.4. MA(1)

136

7.5.5. Asymptotically valid inferences with autocorrelation of unknown form 138

CONTENTS

6

7.5.6. Testing for autocorrelation

142

7.5.7. Lagged dependent variables and autocorrelation

144

7.5.8. Examples

145

Exercises

150

Exercises

152

Chapter 8.

Stochastic regressors

153

8.1. Case 1

154

8.2. Case 2

155

8.3. Case 3

157

8.4. When are the assumptions reasonable?

157

Exercises

160

Chapter 9.

Data problems

161

9.1. Collinearity

161

9.1.1. A brief aside on dummy variables

162

9.1.2. Back to collinearity

163

9.1.3. Detection of collinearity

166

9.1.4. Dealing with collinearity

166

9.2. Measurement error

170

9.2.1. Error of measurement of the dependent variable

170

9.2.2. Error of measurement of the regressors

171

9.3. Missing observations

173

9.3.1. Missing observations on the dependent variable

173

9.3.2. The sample selection problem

176

9.3.3. Missing observations on the regressors

177

Exercises

179

CONTENTS

7

Exercises

179

Exercises

179

Chapter 10. Functional form and nonnested tests

180

10.1. Flexible functional forms

181

10.1.1. The translog form

183

10.1.2. FGLS estimation of a translog model

188

10.2. Testing nonnested hypotheses

192

Chapter 11. Exogeneity and simultaneity

196

11.1. Simultaneous equations

196

11.2. Exogeneity

199

11.3. Reduced form

202

11.4. IV estimation

204

11.5. Identification by exclusion restrictions

210

11.5.1. Necessary conditions

210

11.5.2. Sufficient conditions

213

11.5.3. Example: Klein’s Model 1

219

11.6. 2SLS

222

11.7. Testing the overidentifying restrictions

225

11.8. System methods of estimation

230

11.8.1. 3SLS

232

11.8.2. FIML

237

11.9. Example: 2SLS and Klein’s Model 1

239

Chapter 12. Introduction to the second half

242

Chapter 13. Numeric optimization methods

251

CONTENTS

8

13.1. Search

252

13.2. Derivative-based methods

252

13.2.1. Introduction

252

13.2.2. Steepest descent

254

13.2.3. Newton-Raphson

255

13.3. Simulated Annealing

261

13.4. Examples

261

13.4.1. Discrete Choice: The logit model

261

13.4.2. Count Data: The Poisson model

263

13.4.3. Duration data and the Weibull model

266

13.5. Numeric optimization: pitfalls

270

13.5.1. Poor scaling of the data

270

13.5.2. Multiple optima

271

Exercises

275

Chapter 14. Asymptotic properties of extremum estimators

276

14.1. Extremum estimators

276

14.2. Consistency

277

14.3. Example: Consistency of Least Squares

282

14.4. Asymptotic Normality

283

14.5. Examples

286

14.5.1. Binary response models

286

14.5.2. Example: Linearization of a nonlinear model

290

Chapter 15. Generalized method of moments (GMM)

295

15.1. Definition

295

15.2. Consistency

298

CONTENTS

9


299

15.4. Choosing the weighting matrix

301

15.5. Estimation of the variance-covariance matrix

304

15.5.1. Newey-West covariance estimator

306

15.6. Estimation using conditional moments

307

15.7. Estimation using dynamic moment conditions

312

15.8. A specification test

312

15.9. Other estimators interpreted as GMM estimators

315

15.9.1. OLS with heteroscedasticity of unknown form

315

15.9.2. Weighted Least Squares

317

15.9.3. 2SLS

318

15.9.4. Nonlinear simultaneous equations

319

15.9.5. Maximum likelihood

320

15.10. Example: The Hausman Test

323

15.11. Application: Nonlinear rational expectations

330

15.12. Empirical example: a portfolio model

335

Chapter 16. Quasi-ML

338

16.1. Consistent Estimation of Variance Components

340

16.2. Example: the MEPS Data

342

16.2.1. Infinite mixture models: the negative binomial model

343

16.2.2. Finite mixture models: the mixed negative binomial model

348

16.2.3. Information criteria

350

Exercises

353

Chapter 17. Nonlinear least squares (NLS) 17.1. Introduction and definition

355 355

CONTENTS

10

17.2. Identification

357

17.3. Consistency

359


359

17.5. Example: The Poisson model for count data

361

17.6. The Gauss-Newton algorithm

362

17.7. Application: Limited dependent variables and sample selection

365

17.7.1. Example: Labor Supply

365

Chapter 18. Nonparametric inference

369

18.1. Possible pitfalls of parametric inference: estimation

369

18.2. Possible pitfalls of parametric inference: hypothesis testing

374

18.3. The Fourier functional form

376

18.3.1. Sobolev norm

380

18.3.2. Compactness

381

18.3.3. The estimation space and the estimation subspace

381

18.3.4. Denseness

382

18.3.5. Uniform convergence

383

18.3.6. Identification

384

18.3.7. Review of concepts

384

18.3.8. Discussion

385

18.4. Kernel regression estimators

386

18.4.1. Estimation of the denominator

387

18.4.2. Estimation of the numerator

390

18.4.3. Discussion

391

18.4.4. Choice of the window width: Cross-validation

392

18.5. Kernel density estimation

392

18.6. Semi-nonparametric maximum likelihood

393

CONTENTS

11

18.7. Examples

397

18.7.1. Kernel regression estimation

397

18.7.2. Seminonparametric ML estimation and the MEPS data

398

Chapter 19. Simulation-based estimation

401

19.1. Motivation

401

19.1.1. Example: Multinomial and/or dynamic discrete response models

401

19.1.2. Example: Marginalization of latent variables

404

19.1.3. Estimation of models specified in terms of stochastic differential equations

405

19.2. Simulated maximum likelihood (SML)

407

19.2.1. Example: multinomial probit

408

19.2.2. Properties

410

19.3. Method of simulated moments (MSM)

411

19.3.1. Properties

412

19.3.2. Comments

413

19.4. Efficient method of moments (EMM)

414

19.4.1. Optimal weighting matrix

416

19.4.2. Asymptotic distribution

419

19.4.3. Diagnotic testing

419

19.5. Examples

420

19.5.1. Estimation of stochastic differential equations

420

19.5.2. EMM estimation of a discrete choice model

422

Chapter 20. Parallel programming for econometrics

426

20.1. Example problems

427

20.1.1. Monte Carlo

427

CONTENTS

12

20.1.2. ML

428

20.1.3. GMM

429

20.1.4. Kernel regression

431

Bibliography

433

Chapter 21. Final project: econometric estimation of a RBC model

434

21.1. Data

434

21.2. An RBC Model

436

21.3. A reduced form model

437

21.4. Results (I): The score generator

439

21.5. Solving the structural model

439

Bibliography

441

Chapter 22. Introduction to Octave

442

22.1. Getting started

442

22.2. A short introduction

442

22.3. If you’re running a Linux installation...

444

Chapter 23. Notation and Review

446

23.1. Notation for differentiation of vectors and matrices

446

23.2. Convergenge modes

447

Real-valued sequences:

448

Deterministic real-valued functions

448

Stochastic sequences

448

Stochastic functions

450

23.3. Rates of convergence and asymptotic equality

451

Exercises

453

CONTENTS

13

Chapter 24. The GPL

454

Chapter 25. The attic

464

25.1. Hurdle models

464

25.1.1. Finite mixture models

469

25.2. Models for time series data

473

25.2.1. Basic concepts

473

25.2.2. ARMA models

476

Bibliography

489

Index

490

List of Figures 1.2.1

LYX

20

1.2.2

Octave

21

3.2.1

Typical data, Classical Model

28

3.3.1

Example OLS Fit

31

3.3.2

The fit in observation space

31

3.4.1

Detection of influential observations

35

3.5.1

Uncentered R2

37

3.7.1

Unbiasedness of OLS under classical assumptions

41

3.7.2

Biasedness of OLS when an assumption fails

42

3.7.3

Gauss-Markov Result: The OLS estimator

45

3.7.4

Gauss-Markov Result: The split sample estimator

46

6.5.1

Joint and Individual Confidence Regions

98

6.8.1

RTS as a function of firm size

109

7.4.1

Residuals, Nerlove model, sorted by firm size

126

7.5.1

Autocorrelation induced by misspecification

131

7.5.2

Durbin-Watson critical values

144

7.6.1

Residuals of simple Nerlove model

146

7.6.2

OLS residuals, Klein consumption equation

148

14

LIST OF FIGURES

15

9.1.1

s(β) when there is no collinearity

163

9.1.2

s(β) when there is collinearity

164

9.3.1

Sample selection bias

177

13.1.1

The search method

253

13.2.1

Increasing directions of search

255

13.2.2

Newton-Raphson method

257

13.2.3

Using MuPAD to get analytic derivatives

260

13.4.1

Life expectancy of mongooses, Weibull model

268

13.4.2

Life expectancy of mongooses, mixed Weibull model

270

13.5.1

A foggy mountain

272

15.10.1

OLS

325

15.10.2

IV

325

18.1.1

True and simple approximating functions

371

18.1.2

True and approximating elasticities

372

18.1.3

True function and more flexible approximation

373

18.1.4

True elasticity and more flexible approximation

374

18.6.1

Negative binomial raw moments

396

18.7.1

Kernel fitted OBDV usage versus AGE

398

20.1.1

Speedups from parallelization

432

21.1.1

Consumption and Investment, Levels

434

21.1.2

Consumption and Investment, Growth Rates

435

21.1.3

Consumption and Investment, Bandpass Filtered

435

LIST OF FIGURES

22.2.1

Running an Octave program

16

443

List of Tables 1

Marginal Variances, Sample and Estimated (Poisson)

342

2

Marginal Variances, Sample and Estimated (NB-II)

348

3

Information Criteria, OBDV

351

1

Actual and Poisson fitted frequencies

464

2

Actual and Hurdle Poisson fitted frequencies

469

17

CHAPTER 1

About this document This document integrates lecture notes for a one year graduate level course with computer programs that illustrate and apply the methods that are studied. The immediate availability of executable (and modifiable) example programs when using the PDF1 version of the document is one of the advantages of the system that has been used. On the other hand, when viewed in printed form, the document is a somewhat terse approximation to a textbook. These notes are not intended to be a perfect substitute for a printed textbook. If you are a student of mine, please note that last sentence carefully. There are many good textbooks available. A few of my favorites are listed in the bibliography. With respect to contents, the emphasis is on estimation and inference within the world of stationary data, with a bias toward microeconometrics. The second half is somewhat more polished than the first half, since I have taught that course more often. If you take a moment to read the licensing information in the next section, you’ll see that you are free to copy and modify the document. If anyone would like to contribute material that expands the contents, it would be very welcome. Error corrections and other additions are also welcome. As an example of a project that has made use of these notes, see these very nice lecture slides.

1It is possible to have the program links open up in an editor, ready to run using keyboard macros. To

do this with the PDF version you need to do some setup work. See the bootable CD described below. 18

1.2. OBTAINING THE MATERIALS

19

1.1. License All materials are copyrighted by Michael Creel with the date that appears above. They are provided under the terms of the GNU General Public License, which forms Section 24 of the notes. The main thing you need to know is that you are free to modify and distribute these materials in any way you like, as long as you do so under the terms of the GPL. In particular, you must make available the source files, in editable form, for your modified version of the materials. 1.2. Obtaining the materials The materials are available on my web page, in a variety of forms including PDF and the editable sources, at pareto.uab.es/mcreel/Econometrics/. In addition to the final product, which you’re looking at in some form now, you can obtain the editable sources, which will allow you to create your own version, if you like, or send error corrections and contributions. The main document was prepared using LYX (www.lyx.org) and Octave (www.octave.org). LYX is a free2 “what you see is what you mean” word processor, basically working as a graphical frontend to LATEX. It (with help from other applications) can export your work in LATEX, HTML, PDF and several other forms. It will run on Linux, Windows, and MacOS systems. Figure 1.2.1 shows LYX editing this document. GNU Octave has been used for the example programs, which are scattered though the document. This choice is motivated by two factors. The first is the high quality of the Octave environment for doing applied econometrics. The fundamental tools exist and are implemented in a way that make extending them fairly easy. The example programs included here may convince you of this point. Secondly, Octave’s licensing philosophy fits in with the goals of this project. Thirdly, it runs on Linux, Windows 2”Free” is used in the sense of ”freedom”, but L X is also free of charge. Y

1.3. AN EASY WAY TO USE LYX AND OCTAVE TODAY

20

F IGURE 1.2.1. LYX

and MacOS. Figure 1.2.2 shows an Octave program being edited by NEdit, and the result of running the program in a shell window.

1.3. An easy way to use LYX and Octave today The example programs are available as links to files on my web page in the PDF version, and here. Support files needed to run these are available here. The files won’t run properly from your browser, since there are dependencies between files - they are only illustrative when browsing. To see how to use these files (edit and run them), you should go to the home page of this document, since you will probably want to download the pdf version together with all the support files and examples. Then set

1.3. AN EASY WAY TO USE LYX AND OCTAVE TODAY

21

F IGURE 1.2.2. Octave

the base URL of the PDF file to point to wherever the Octave files are installed. All of this may sound a bit complicated, because it is. An easier solution is available: The file pareto.uab.es/mcreel/Econometrics/econometrics.iso distribution of Linux is an ISO image file that may be burnt to CDROM. It contains a bootable-from-CD Gnu/Linux system that has all of the tools needed to edit this document, run the Octave example programs, etcetera. In particular, it will allow you to cut out small portions of the notes and edit them, and send them to me as LYX (or TEX) files for inclusion in future versions. Think error corrections, additions, etc.! The CD automatically detects the hardware of your computer, and will not touch your hard disk unless you explicitly tell it to do so. It is based upon the ParallelKnoppix GNU/Linux distribution. The

1.4. KNOWN BUGS

22

reason why these notes are integrated into a Linux distribution for parallel computing will be apparent if you get to Chapter 20. 1.4. Known Bugs This section is a reminder to myself to try to fix a few things. • The PDF version has hyperlinks to figures that jump to the wrong figure. The numbers are correct, but the links are not. ps2pdf bugs?

CHAPTER 2

Introduction: Economic and econometric models Economic theory tells us that the demand function for a good is something like: x = x(p, m, z) • x is the quantity demanded • p is G × 1 vector of prices of the good and its substitutes and complements • m is income • z is a vector of other variables such as individual characteristics that affect preferences

Suppose we have a sample consisting of one observation on n individuals’ demands at time period t (this is a cross section, where i = 1, 2, ..., n indexes the individuals in the sample). The individual demand functions are xi = xi (pi , mi , zi ) The model is not estimable as it stands, since:

• The form of the demand function is different for all i. • Some components of zi may not be observable to an outside modeler. For example, people don’t eat the same lunch every day, and you can’t tell what they will order just by looking at them. Suppose we can break zi into the observable components wi and a single unobservable component εi . 23

2. INTRODUCTION: ECONOMIC AND ECONOMETRIC MODELS

24

A step toward an estimable econometric model is to suppose that the model may be written as xi = β1 + p0i β p + mi βm + w0i βw + εi We have imposed a number of restrictions on the theoretical model: • The functions xi (·) which in principle may differ for all i have been restricted to all belong to the same parametric family. • Of all parametric families of functions, we have restricted the model to the class of linear in the variables functions. • The parameters are constant across individuals. • There is a single unobservable component, and we assume it is additive. If we assume nothing about the error term ε, we can always write the last equation. But in order for the β coefficients to have an economic meaning, and in order to be able to estimate them from sample data, we need to make additional assumptions. These additional assumptions have no theoretical basis, they are assumptions on top of those needed to prove the existence of a demand function. The validity of any results we obtain using this model will be contingent on these additional restrictions being at least approximately correct. For this reason, specification testing will be needed, to check that the model seems to be reasonable. Only when we are convinced that the model is at least approximately correct should we use it for economic analysis. When testing a hypothesis using an econometric model, three factors can cause a statistical test to reject the null hypothesis: (1) the hypothesis is false (2) a type I error has occured (3) the econometric model is not correctly specified so the test does not have the assumed distribution

2. INTRODUCTION: ECONOMIC AND ECONOMETRIC MODELS

25

We would like to ensure that the third reason is not contributing to rejections, so that rejection will be due to either the first or second reasons. Hopefully the above example makes it clear that there are many possible sources of misspecification of econometric models. In the next few sections we will obtain results supposing that the econometric model is entirely correctly specified. Later we will examine the consequences of misspecification and see some methods for determining if a model is correctly specified. Later on, econometric methods that seek to minimize maintained assumptions are introduced.

CHAPTER 3

Ordinary Least Squares 3.1. The Linear Model Consider approximating a variable y using the variables x1 , x2 , ..., xk . We can consider a model that is a linear approximation: Linearity: the model is a linear function of the parameter vector β0 : y = β01 x1 + β02 x2 + ... + β0k xk + ε or, using vector notation: y = x 0 β0 + ε The dependent variable y is a scalar random variable, x = ( x1 x2 · · · xk )0 is a kvector of explanatory variables, and β0 = ( β01 β02 · · · β0k ) . The superscript “0” in 0

β0 means this is the ”true value” of the unknown parameter. It will be defined more precisely later, and usually suppressed when it’s not necessary for clarity.

Suppose that we want to use data to try to determine the best linear approximation to y using the variables x. The data {(yt , xt )} ,t = 1, 2, ..., n are obtained by some form of sampling1. An individual observation is thus

yt = xt0 β + εt

1For example, cross-sectional data may be obtained by random sampling. Time series data accumulate

historically.

26

3.2. ESTIMATION BY LEAST SQUARES

27

The n observations can be written in matrix form as y = Xβ + ε,

(3.1.1) where y =

y1 y2 · · · yn

0

is n × 1 and X =

x1 x2 · · · xn

0

.

Linear models are more general than they might first appear, since one can employ nonlinear transformations of the variables: ϕ0 (z) =

h

ϕ1 (w) ϕ2 (w) · · · ϕ p (w)

i

β+ε

where the φi () are known functions. Defining y = ϕ0 (z), x1 = ϕ1 (w), etc. leads to a model in the form of equation 3.6.1. For example, the Cobb-Douglas model β

β

z = Aw2 2 w3 3 exp(ε) can be transformed logarithmically to obtain ln z = ln A + β2 ln w2 + β3 ln w3 + ε. If we define y = ln z, β1 = ln A,etc., we can put the model in the form needed. The approximation is linear in the parameters, but not necessarily linear in the variables.

3.2. Estimation by least squares Figure 3.2.1, obtained by running TypicalData.m shows some data that follows the linear model yt = β1 +β2 xt2 +εt . The green line is the ”true” regression line β1 +β2 xt2 , and the red crosses are the data points (xt2 , yt ), where εt is a random error that has mean zero and is independent of xt2 . Exactly how the green line is defined will become clear later. In practice, we only have the data, and we don’t know where the green line lies. We need to gain information about the straight line that best fits the data points.


28

F IGURE 3.2.1. Typical data, Classical Model 10

data true regression line

5

0

-5

-10

-15

0

2

4

6

8

10 X

12

14

16

18

20

The ordinary least squares (OLS) estimator is defined as the value that minimizes the sum of the squared errors: βˆ = arg min s(β) where

s(β) =

n

∑

t=1

yt − xt0 β

2

= (y − Xβ)0 (y − Xβ) = y0 y − 2y0 Xβ + β0 X0 Xβ = k y − Xβ k2


29

This last expression makes it clear how the OLS estimator is defined: it minimizes the Euclidean distance between y and X β. The fitted OLS coefficients will define the best linear approximation to y using x as basis functions, where ”best” means minimum Euclidean distance. One could think of other estimators based upon other metrics. For n example, the minimum absolute distance (MAD) minimizes ∑t=1 |yt − xt0 β|. Later, we

will see that which estimator is best in terms of their statistical properties, rather than in terms of the metrics that define them, depends upon the properties of ε, about which we have as yet made no assumptions. • To minimize the criterion s(β), find the derivative with respect to β and it to zero: ˆ = −2X0 y + 2X0 Xβˆ = 0 Dβ s(β) so βˆ = (X0 X)−1 X0 y. • To verify that this is a minimum, check the s.o.s.c.: ˆ = 2X0 X D2β s(β) Since ρ(X) = K, this matrix is positive definite, since it’s a quadratic form in a p.d. matrix (identity matrix of order n), so βˆ is in fact a minimizer. ˆ • The fitted values are in the vector yˆ = Xβ.

• The residuals are in the vector εˆ = y − Xβˆ • Note that y = Xβ + ε = Xβˆ + εˆ

3.3. GEOMETRIC INTERPRETATION OF LEAST SQUARES ESTIMATION

30

• Also, the first order conditions can be written as X0 y − X0 Xβˆ = 0 0 ˆ X y − Xβ = 0 X0 εˆ = 0

which is to say, the OLS residuals are orthogonal to X. Let’s look at this more carefully.

3.3. Geometric interpretation of least squares estimation 3.3.1. In X ,Y Space. Figure 3.3.1 shows a typical fit to data, along with the true regression line. Note that the true line and the estimated line are different. This figure was created by running the Octave program OlsFit.m . You can experiment with changing the parameter values to see how this affects the fit, and to see how the fitted line will sometimes be close to the true line, and sometimes rather far away. 3.3.2. In Observation Space. If we want to plot in observation space, we’ll need to use only two or three observations, or we’ll encounter some limitations of the blackboard. Let’s use two. With only two observations, we can’t have K > 1. • We can decompose y into two components: the orthogonal projection onto

ˆ and the component that is the the K−dimensional space spanned by X , X β,

orthogonal projection onto the n − K subpace that is orthogonal to the span of X , εˆ .

• Since βˆ is chosen to make εˆ as short as possible, εˆ will be orthogonal to the

space spanned by X . Since X is in this space, X 0εˆ = 0. Note that the f.o.c. that define the least squares estimator imply that this is so.


31

F IGURE 3.3.1. Example OLS Fit 15

data points fitted line true line

10

5

0

-5

-10

-15

0

2

4

6

8

10 X

12

14

16

18

20

F IGURE 3.3.2. The fit in observation space Observation 2

y

e = M_xY

S(x)

x x*beta=P_xY

Observation 1


32

3.3.3. Projection Matrices. X βˆ is the projection of y onto the span of X , or X βˆ = X X 0 X

−1

X 0y

Therefore, the matrix that projects y onto the span of X is PX = X (X 0X )−1 X 0 since X βˆ = PX y. εˆ is the projection of y onto the N − K dimensional space that is orthogonal to the span of X . We have that εˆ = y − X βˆ = y − X (X 0X )−1 X 0 y =

In − X (X 0X )−1 X 0 y.

So the matrix that projects y onto the space orthogonal to the span of X is MX = In − X (X 0X )−1 X 0 = In − PX . We have εˆ = MX y. Therefore y = PX y + MX y = X βˆ + εˆ .

3.4. INFLUENTIAL OBSERVATIONS AND OUTLIERS

33

These two projection matrices decompose the n dimensional vector y into two orthogonal components - the portion that lies in the K dimensional space defined by X , and the portion that lies in the orthogonal n − K dimensional space. • Note that both PX and MX are symmetric and idempotent. – A symmetric matrix A is one such that A = A0 . – An idempotent matrix A is one such that A = AA. – The only nonsingular idempotent matrix is the identity matrix.

3.4. Influential observations and outliers The OLS estimator of the ith element of the vector β0 is simply βˆ i =

(X 0X )−1 X 0

= c0i y

i·

y

This is how we define a linear estimator - it’s a linear function of the dependent variable. Since it’s a linear combination of the observations on the dependent variable, where the weights are detemined by the observations on the regressors, some observations may have more influence than others. Define ht = (PX )tt = et0 PX et = k PX et k2 ≤ k et k2 = 1

3.4. INFLUENTIAL OBSERVATIONS AND OUTLIERS

34

ht is the tth element on the main diagonal of PX ( et is a n vector of zeros with a 1 in the tth position). So 0 < ht < 1, and TrPX = K ⇒ h = K/n. So, on average, the weight on the yt ’s is K/n. If the weight is much higher, then the observation has the potential to affect the fit importantly. The weight, ht is referred to as the leverage of the observation. However, an observation may also be influential due to the value of yt , rather than the weight it is multiplied by, which only depends on the xt ’s. To account for this, consider estimation of β without using the t th observation (designate this estimator as βˆ (t) ). One can show (see Davidson and MacKinnon, pp. 32-5 for proof) that βˆ (t) = βˆ −

(X 0 X )−1 Xt0 εˆ t

1 1 − ht

so the change in the t th observations fitted value is Xt βˆ − Xt βˆ (t) =

ht 1 − ht

εˆ t

While an observation may be influential if it doesn’t affect its own fitted value, it certainly is influential if it does. A fast means of identifying influential observations ht is to plot 1−h εˆ t (which I will refer to as the own influence of the observation) as a t function of t. Figure 3.4.1 gives an example plot of data, fit, leverage and influence.

The Octave program is InfluentialObservation.m . If you re-run the program you will see that the leverage of the last observation (an outlying value of x) is always high, and the influence is sometimes high. After influential observations are detected, one needs to determine why they are influential. Possible causes include:

3.5. GOODNESS OF FIT

35

F IGURE 3.4.1. Detection of influential observations 14

Data points fitted Leverage Influence

12

10

8

6

4

2

0

-2

0

0.5

1

1.5 X

2

2.5

3

• data entry error, which can easily be corrected once detected. Data entry errors are very common. • special economic factors that affect some observations. These would need to be identified and incorporated in the model. This is the idea behind structural change: the parameters may not be constant across all observations. • pure randomness may have caused us to sample a low-probability observation. There exist robust estimation methods that downweight outliers.

3.5. Goodness of fit The fitted model is y = X βˆ + εˆ


36

Take the inner product: y0 y = βˆ 0 X 0 X βˆ + 2βˆ 0 X 0 εˆ + εˆ 0 εˆ But the middle term of the RHS is zero since X 0 εˆ = 0, so y0 y = βˆ 0 X 0X βˆ + εˆ 0 εˆ

(3.5.1) The uncentered R2u is defined as

εˆ 0 εˆ y0 y βˆ 0 X 0 X βˆ

R2u = 1 − = =

y0 y k PX y k2 k y k2

= cos2 (φ), where φ is the angle between y and the span of X .

• The uncentered R2 changes if we add a constant to y, since this changes φ (see Figure 3.5.1, the yellow vector is a constant, since it’s on the 45 degree line in observation space). Another, more common definition measures the contribution of the variables, other than the constant term, to explaining the variation in y. Thus it measures the ability of the model to explain the variation of y about its unconditional sample mean.

Let ι = (1, 1, ..., 1)0, a n -vector. So Mι = In − ι(ι0 ι)−1 ι0 = In − ιι0 /n


37

F IGURE 3.5.1. Uncentered R2

Mι y just returns the vector of deviations from the mean. In terms of deviations from the mean, equation 3.5.1 becomes y0 Mι y = βˆ 0 X 0 Mι X βˆ + εˆ 0 Mι εˆ The centered R2c is defined as R2c = 1 −

εˆ 0 εˆ ESS = 1− 0 y Mι y T SS

n (yt − y) ¯ 2. where ESS = εˆ 0 εˆ and T SS = y0 Mι y=∑t=1

Supposing that X contains a column of ones (i.e., there is a constant term), X 0 εˆ = 0 ⇒ ∑ εˆ t = 0 t

3.6. THE CLASSICAL LINEAR REGRESSION MODEL

38

so Mι εˆ = εˆ . In this case y0 Mι y = βˆ 0 X 0Mι X βˆ + εˆ 0 εˆ So R2c =

RSS T SS

where RSS = βˆ 0 X 0 Mι X βˆ • Supposing that a column of ones is in the space spanned by X (PX ι = ι), then one can show that 0 ≤ R2c ≤ 1.

3.6. The classical linear regression model Up to this point the model is empty of content beyond the definition of a best linear approximation to y and some geometrical properties. There is no economic content to the model, and the regression parameters have no economic interpretation. For example, what is the partial derivative of y with respect to x j ? The linear approximation is y = β1 x1 + β2 x2 + ... + βk xk + ε The partial derivative is

∂y ∂ε = βj + ∂x j ∂x j

Up to now, there’s no guarantee that

∂ε ∂x j =0.

For the β to have an economic meaning, we

need to make additional assumptions. The assumptions that are appropriate to make depend on the data under consideration. We’ll start with the classical linear regression model, which incorporates some assumptions that are clearly not realistic for economic data. This is to be able to explain some concepts with a minimum of confusion and notational clutter. Later we’ll adapt the results to what we can get with more realistic assumptions.

3.6. THE CLASSICAL LINEAR REGRESSION MODEL

39

Linearity: the model is a linear function of the parameter vector β0 : (3.6.1)

y = β01 x1 + β02 x2 + ... + β0k xk + ε

or, using vector notation: y = x 0 β0 + ε Nonstochastic linearly independent regressors: X is a fixed matrix of constants, it has rank K, its number of columns, and 3.6.2 (3.6.2)

1 lim X0 X = QX n

where QX is a finite positive definite matrix. This is needed to be able to identify the individual effects of the explanatory variables. Independently and identically distributed errors: (3.6.3)

ε ∼ IID(0, σ2In )

ε is jointly distributed IIN. This implies the following two properties: Homoscedastic errors: (3.6.4)

V (εt ) = σ20 , ∀t

Nonautocorrelated errors:

(3.6.5)

E (εt εs ) = 0, ∀t 6= s

Optionally, we will sometimes assume that the errors are normally distributed.

3.7. SMALL SAMPLE STATISTICAL PROPERTIES OF THE LEAST SQUARES ESTIMATOR

40

Normally distributed errors: (3.6.6)

ε ∼ N(0, σ2 In )

3.7. Small sample statistical properties of the least squares estimator Up to now, we have only examined numeric properties of the OLS estimator, that always hold. Now we will examine statistical properties. The statistical properties depend upon the assumptions we can make. 3.7.1. Unbiasedness. We have βˆ = (X 0 X )−1 X 0 y. By linearity, βˆ = (X 0 X )−1 X 0 (X β + ε) = β + (X 0X )−1 X 0 ε By 3.6.2 and 3.6.3 E(X 0 X )−1 X 0 ε = E(X 0X )−1 X 0 ε = (X 0X )−1 X 0 Eε = 0 so the OLS estimator is unbiased under the assumptions of the classical model. Figure 3.7.1 shows the results of a small Monte Carlo experiment where the OLS estimator was calculated for 10000 samples from the classical model with y = 1 + 2x + ε, where n = 20, σ2ε = 9, and x is fixed across samples. We can see that the β2 appears to be estimated without bias. The program that generates the plot is Unbiased.m , if you would like to experiment with this. With time series data, the OLS estimator will often be biased. Figure 3.7.2 shows the results of a small Monte Carlo experiment where the OLS estimator was calculated


41

F IGURE 3.7.1. Unbiasedness of OLS under classical assumptions Beta hat - Beta true 0.12

0.1

0.08

0.06

0.04

0.02

0

-3

-2

-1

0

1

2

3

for 1000 samples from the AR(1) model with yt = 0 + 0.9yt−1 + εt , where n = 20 and σ2ε = 1. In this case, assumption 3.6.2 does not hold: the regressors are stochastic. We can see that the bias in the estimation of β2 is about -0.2. The program that generates the plot is Biased.m , if you would like to experiment with this. 3.7.2. Normality. With the linearity assumption, we have βˆ = β + (X 0X )−1 X 0 ε. This is a linear function of ε. Adding the assumption of normality (3.6.6, which implies strong exogeneity), then βˆ ∼ N β, (X 0X )−1 σ20

since a linear function of a normal random vector is also normally distributed. In Figure 3.7.1 you can see that the estimator appears to be normally distributed. It in fact is normally distributed, since the DGP (see the Octave program) has normal errors.


42

F IGURE 3.7.2. Biasedness of OLS when an assumption fails Beta hat - Beta true 0.14

0.12

0.1

0.08

0.06

0.04

0.02

0 -1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

Even when the data may be taken to be IID, the assumption of normality is often questionable or simply untenable. For example, if the dependent variable is the number of automobile trips per week, it is a count variable with a discrete distribution, and is thus not normally distributed. Many variables in economics can take on only nonnegative values, which, strictly speaking, rules out normality.2

3.7.3. The variance of the OLS estimator and the Gauss-Markov theorem. Now let’s make all the classical assumptions except the assumption of normality. We

2Normality may be a good model nonetheless, as long as the probability of a negative value occuring is

negligable under the model. This depends upon the mean being large enough in relation to the variance.


43

ˆ = β. So have βˆ = β + (X 0X )−1 X 0 ε and we know that E(β) ˆ = E Var(β)

βˆ − β

βˆ − β

0

= E (X 0 X )−1 X 0 εε0 X (X 0X )−1 = (X 0 X )−1 σ20

The OLS estimator is a linear estimator, which means that it is a linear function of the dependent variable, y. βˆ =

(X 0 X )−1 X 0 y

= Cy

where C is a function of the explanatory variables only, not the dependent variable. It is also unbiased under the present assumptions, as we proved above. One could consider other weights W that are a function of X that define some other linear estimator. We’ll still insist upon unbiasedness. Consider β˜ = Wy, where W = W (X ) is some k × n matrix function of X . Note that since W is a function of X , it is nonstochastic, too. If the estimator is unbiased, then we must have W X = IK : E (Wy)

=

E (W X β0 +W ε)

=

W X β0

=

β0

⇒ WX

=

IK

The variance of β˜ is ˜ = WW 0 σ2 . V (β) 0


44

Define D = W − (X 0 X )−1 X 0 so W = D + (X 0 X )−1 X 0 Since W X = IK , DX = 0, so 0 D + (X 0 X )−1 X 0 D + (X 0 X )−1 X 0 σ20 −1 2 = DD0 + X 0 X σ0

˜ = V (β)

So

˜ ≥ V (β) ˆ V (β) ˜ − V (β) ˆ The inequality is a shorthand means of expressing, more formally, that V ( β) is a positive semi-definite matrix. This is a proof of the Gauss-Markov Theorem. The OLS estimator is the ”best linear unbiased estimator” (BLUE).

• It is worth emphasizing again that we have not used the normality assumption in any way to prove the Gauss-Markov theorem, so it is valid if the errors are not normally distributed, as long as the other assumptions hold.

To illustrate the Gauss-Markov result, consider the estimator that results from splitting the sample into p equally-sized parts, estimating using each part of the data separately by OLS, then averaging the p resulting estimators. You should be able to show that this estimator is unbiased, but inefficient with respect to the OLS estimator. The program Efficiency.m illustrates this using a small Monte Carlo experiment, which compares the OLS estimator and a 3-way split sample estimator. The data generating process follows the classical model, with n = 21. The true parameter value is β = 2. In Figures


45

F IGURE 3.7.3. Gauss-Markov Result: The OLS estimator Beta 2 hat, OLS 0.1 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0

0

0.5

1

1.5

2

2.5

3

3.5

4

3.7.3 and 3.7.4 we can see that the OLS estimator is more efficient, since the tails of its histogram are more narrow.

−1 ˆ = β and Var(β) ˆ = X 0X We have that E(β) σ20 , but we still need to estimate

the variance of ε, σ20 , in order to have an idea of the precision of the estimates of β. A commonly used estimator of σ20 is

This estimator is unbiased:

c2 = σ 0

1 0 εˆ εˆ n−K


F IGURE 3.7.4. Gauss-Markov Result: The split sample estimator Beta 2 hat, Split Sample Estimator 0.12

0.1

0.08

0.06

0.04

0.02

0

0

0.5

1

1.5

c2 = σ 0

=

c2 ) = E (σ 0

=

= = =

2

2.5

1 0 εˆ εˆ n−K 1 0 ε Mε n−K 1 E(Trε0 Mε) n−K 1 E(TrMεε0) n−K 1 TrEX Eε|X (Mεε0 ) n−K 1 σ20 EX TrM n−K 1 σ2 (n − k) n−K 0

= σ20

3

3.5

4

46

3.8. EXAMPLE: THE NERLOVE MODEL

47

where we use the fact that Tr(AB) = Tr(BA) when both products are conformable. Thus, this estimator is also unbiased under these assumptions.

3.8. Example: The Nerlove model 3.8.1. Theoretical background. For a firm that takes input prices w and the output level q as given, the cost minimization problem is to choose the quantities of inputs x to solve the problem

min w0 x x

subject to the restriction

f (x) = q. The solution is the vector of factor demands x(w, q). The cost function is obtained by substituting the factor demands into the criterion function: Cw, q) = w0 x(w, q).

• Monotonicity Increasing factor prices cannot decrease cost, so ∂C(w, q) ≥0 ∂w Remember that these derivatives give the conditional factor demands (Shephard’s Lemma). • Homogeneity The cost function is homogeneous of degree 1 in input prices: C(tw, q) = tC(w, q) where t is a scalar constant. This is because the factor


48

demands are homogeneous of degree zero in factor prices - they only depend upon relative prices. • Returns to scale The returns to scale parameter γ is defined as the inverse of the elasticity of cost with respect to output: γ=

∂C(w, q) q ∂q C(w, q)

−1

Constant returns to scale is the case where increasing production q implies that cost increases in the proportion 1:1. If this is the case, then γ = 1.

3.8.2. Cobb-Douglas functional form. The Cobb-Douglas functional form is linear in the logarithms of the regressors and the dependent variable. For a cost function, if there are g factors, the Cobb-Douglas cost function has the form β

β

C = Aw1 1 ...wg g qβq eε What is the elasticity of C with respect to w j ? eCw j

=

∂C ∂WJ

β

wj C

β −1

= β j Aw1 1 .w j j

wj

β

..wg g qβq eε

β

β

Aw1 1 ...wg g qβq eε

= βj This is one of the reasons the Cobb-Douglas form is popular - the coefficients are easy to interpret, since they are the elasticities of the dependent variable with respect to the


49

explanatory variable. Not that in this case, eCw j

=

∂C ∂WJ

= x j (w, q)

wj C

wj C

≡ s j (w, q) the cost share of the jth input. So with a Cobb-Douglas cost function, β j = s j (w, q). The cost shares are constants. Note that after a logarithmic transformation we obtain lnC = α + β1 ln w1 + ... + βg ln wg + βq ln q + ε where α = ln A . So we see that the transformed model is linear in the logs of the data. One can verify that the property of HOD1 implies that g

∑ βg = 1

i=1

In other words, the cost shares add up to 1. The hypothesis that the technology exhibits CRTS implies that γ=

1 =1 βq

so βq = 1. Likewise, monotonicity implies that the coefficients βi ≥ 0, i = 1, ..., g.

3.8.3. The Nerlove data and OLS. The file nerlove.data contains data on 145 electric utility companies’ cost of production, output and input prices. The data are for the U.S., and were collected by M. Nerlove. The observations are by row, and the columns are COMPANY, COST (C), OUTPUT (Q), PRICE OF LABOR (PL ),


50

PRICE OF FUEL (PF ) and PRICE OF CAPITAL (PK ). Note that the data are sorted by output level (the third column). We will estimate the Cobb-Douglas model lnC = β1 + β2 ln Q + β3 ln PL + β4 ln PF + β5 ln PK + ε

(3.8.1)

using OLS. To do this yourself, you need the data file mentioned above, as well as Nerlove.m (the estimation program) , and the library of Octave functions mentioned in the introduction to Octave that forms section 22 of this document.3 The results are ********************************************************* OLS estimation results Observations 145 R-squared 0.925955 Sigma-squared 0.153943 Results (Ordinary var-cov estimator) constant output labor fuel capital

estimate -3.527 0.720 0.436 0.427 -0.220

st.err. 1.774 0.017 0.291 0.100 0.339

t-stat. -1.987 41.244 1.499 4.249 -0.648

p-value 0.049 0.000 0.136 0.000 0.518

********************************************************* • Do the theoretical restrictions hold? • Does the model fit well? • What do you think about RTS? While we will use Octave programs as examples in this document, since following the programming statements is a useful way of learning how theory is put into practice, 3If you are running the bootable CD, you have all of this installed and ready to run.


51

you may be interested in a more ”user-friendly” environment for doing econometrics. I heartily recommend Gretl, the Gnu Regression, Econometrics, and Time-Series Library. This is an easy to use program, available in English, French, and Spanish, and it comes with a lot of data ready to use. It even has an option to save output as LATEX fragments, so that I can just include the results into this document, no muss, no fuss. Here the results of the Nerlove model from GRETL:

Model 2: OLS estimates using the 145 observations 1–145 Dependent variable: l_cost Variable const

Coefficient

Std. Error

−3.5265

t-statistic

p-value

1.77437

−1.9875

0.0488

l_output

0.720394

0.0174664

41.2445

0.0000

l_labor

0.436341

0.291048

1.4992

0.1361

l_fuel

0.426517

0.100369

4.2495

0.0000

−0.219888

0.339429

−0.6478

0.5182

l_capita

Mean of dependent variable

1.72466

S.D. of dependent variable

1.42172

Sum of squared residuals

21.5520

ˆ Standard error of residuals (σ)

0.392356

Unadjusted R2

0.925955

Adjusted R¯ 2

0.923840

F(4, 140)

437.686

Akaike information criterion

145.084

Schwarz Bayesian criterion

159.967


52

Fortunately, Gretl and my OLS program agree upon the results. Gretl is included in the bootable CD mentioned in the introduction. I recommend using GRETL to repeat the examples that are done using Octave. The previous properties hold for finite sample sizes. Before considering the asymptotic properties of the OLS estimator it is useful to review the MLE estimator, since under the assumption of normal errors the two estimators coincide.

EXERCISES

53

Exercises (1) Prove that the split sample estimator used to generate figure 3.7.4 is unbiased. (2) Calculate the OLS estimates of the Nerlove model using Octave and GRETL, and provide printouts of the results. Interpret the results. (3) Do an analysis of whether or not there are influential observations for OLS estimation of the Nerlove model. Discuss. (4) Using GRETL, examine the residuals after OLS estimation and tell me whether or not you believe that the assumption of independent identically distributed normal errors is warranted. No need to do formal tests, just look at the plots. Print out any that you think are relevant, and interpret them. (5) For a random vector X ∼ N(µx , Σ), what is the distribution of AX + b, where A and b are conformable matrices of contants? (6) Using Octave, write a little program that verifies that Tr(AB) = Tr(BA) for A and B 4x4 matrices of random numbers. Note: there is an Octave function trace. (7) For the model with a constant and a single regressor, yt = β1 + β2 xt + εt , which satisfies the classical assumptions, prove that the variance of the OLS estimator declines to zero as the sample size increases.

CHAPTER 4

Maximum likelihood estimation The maximum likelihood estimator is important since it is asymptotically efficient, as is shown below. For the classical linear model with normal errors, the ML and OLS estimators of β are the same, so the following theory is presented without examples. In the second half of the course, nonlinear models with nonnormal errors are introduced, and examples may be found there.

4.1. The likelihood function Suppose we have a sample of size n of the random vectors y and z. Suppose the joint density of Y = y1 . . . yn and Z = z1 . . . zn is characterized by a

parameter vector ψ0 :

fY Z (Y, Z, ψ0 ). This is the joint density of the sample. This density can be factored as fY Z (Y, Z, ψ0 ) = fY |Z (Y |Z, θ0 ) fZ (Z, ρ0 ) The likelihood function is just this density evaluated at other values ψ L(Y, Z, ψ) = f (Y, Z, ψ), ψ ∈ Ψ, where Ψ is a parameter space. The maximum likelihood estimator of ψ0 is the value of ψ that maximizes the likelihood function. 54

4.1. THE LIKELIHOOD FUNCTION

55

Note that if θ0 and ρ0 share no elements, then the maximizer of the conditional likelihood function fY |Z (Y |Z, θ) with respect to θ is the same as the maximizer of

the overall likelihood function fY Z (Y, Z, ψ) = fY |Z (Y |Z, θ) fZ (Z, ρ), for the elements

of ψ that correspond to θ. In this case, the variables Z are said to be exogenous for

estimation of θ, and we may more conveniently work with the conditional likelihood function fY |Z (Y |Z, θ) for the purposes of estimating θ0 .

D EFINITION 4.1.1. The maximum likelihood estimator of θ0 = arg max fY |Z (Y |Z, θ)

• If the n observations are independent, the likelihood function can be written as

n

L(Y |Z, θ) = ∏ f (yt |zt , θ) t=1

where the ft are possibly of different form. • If this is not possible, we can always factor the likelihood into contributions of observations, by using the fact that a joint density can be factored into the product of a marginal and conditional (doing this iteratively) L(Y, θ) = f (y1 |z1 , θ) f (y2 |y1 , z2 , θ) f (y3 |y1 , y2 , z3 , θ) · · · f (yn |y1, y2 , . . . yt−n , zn , θ) To simplify notation, define xt = {y1 , y2 , ..., yt−1 , zt } so x1 = z1 , x2 = {y1 , z2 }, etc. - it contains exogenous and predetermined endogeous variables. Now the likelihood function can be written as n

L(Y, θ) = ∏ f (yt |xt , θ) t=1


56

The criterion function can be defined as the average log-likelihood function: 1 1 n sn (θ) = ln L(Y, θ) = ∑ ln f (yt |xt , θ) n n t=1 The maximum likelihood estimator may thus be defined equivalently as θˆ = arg max sn (θ), where the set maximized over is defined below. Since ln(·) is a monotonic increasing ˆ function, ln L and L maximize at the same value of θ. Dividing by n has no effect on θ.

4.1.1. Example: Bernoulli trial. Suppose that we are flipping a coin that may be biased, so that the probability of a heads may not be 0.5. Maybe we’re interested in estimating the probability of a heads. Let y = 1(heads) be a binary variable that indicates whether or not a heads is observed. The outcome of a toss is a Bernoulli random variable: y

fY (y, p0 ) = p0 (1 − p0 )1−y , y ∈ {0, 1} = 0, y ∈ / {0, 1} So a representative term that enters the likelihood function is fY (y, p) = py (1 − p)1−y and ln fY (y, p) = y ln p + (1 − y) ln (1 − p)


57

The derivative of this is ∂ ln fY (y, p) = ∂p =

y (1 − y) − p (1 − p) y− p p (1 − p)

Averaging this over a sample of size n gives ∂sn (p) 1 n yi − p = ∑ ∂p n i=1 p (1 − p) Setting to zero and solving gives pˆ = y¯ So it’s easy to calculate the MLE of p0 in this case. Now imagine that we had a bag full of bent coins, each bent around a sphere of a different radius (with the head pointing to the outside of the sphere). We might suspect that the probability of a heads could depend upon the radius. Suppose that h i0 −1 0 pi ≡ p(xi , β) = (1 + exp(−xi β)) where xi = 1 ri , so that β is a 2×1 vector. Now

∂pi (β) = pi (1 − pi ) xi ∂β

so

∂ ln fY (y, β) = ∂β

y − pi pi (1 − pi ) xi pi (1 − pi )

= (yi − p(xi , β)) xi

4.2. CONSISTENCY OF MLE

58

So the derivative of the average log lihelihood function is now ∂sn (β) ∑ni=1 (yi − p(xi , β)) xi = ∂β n This is a set of 2 nolinear equations in the two unknown elements in β. There is no explicit solution for the two elements that set the equations to zero. This is common with ML estimators, they are often nonlinear, and finding their values often require use of numeric methods to find solutions to the first order conditions. 4.2. Consistency of MLE To show consistency of the MLE, we need to make explicit some assumptions. Compact parameter space: θ ∈ Θ, an open bounded subset of ℜK . Maximixation is over Θ, which is compact.

This implies that θ is an interior point of the parameter space Θ. Uniform convergence: u.a.s

sn (θ) → lim Eθ0 sn (θ) ≡ s∞ (θ, θ0 ), ∀θ ∈ Θ. n→∞

We have suppressed Y here for simplicity. This requires that almost sure convergence holds for all possible parameter values. For a given parameter value, an ordinary Law of Large Numbers will usually imply almost sure convergence to the limit of the expectation. Convergence for a single element of the parameter space, combined with the assumption of a compact parameter space, ensures uniform convergence. Continuity: sn (θ) is continuous in θ, θ ∈ Θ. This implies that s∞ (θ, θ0 ) is continuous in θ.

Identification: s∞ (θ, θ0 ) has a unique maximum in its first argument. a.s. We will use these assumptions to show that θˆn → θ0 .

4.2. CONSISTENCY OF MLE

59

First, θˆ n certainly exists, since a continuous function has a maximum on a compact set. Second, for any θ 6= θ0

L(θ) L(θ) E ln ≤ ln E L(θ0 ) L(θ0 )

by Jensen’s inequality ( ln (·) is a concave function). Now, the expectation on the RHS is E

L(θ) L(θ0 )

=

Z

L(θ) L(θ0 )dy = 1, L(θ0 )

since L(θ0 ) is the density function of the observations, and since the integral of any density is 1. Therefore, since ln(1) = 0, L(θ) ≤ 0, E ln L(θ0 ) or E (sn (θ)) − E (sn (θ0 )) ≤ 0.

Taking limits, this is s∞ (θ, θ0 ) − s∞ (θ0 , θ0 ) ≤ 0 except on a set of zero probability (by the uniform convergence assumption). By the identification assumption there is a unique maximizer, the inequality is strict if θ 6= θ0 :

s∞ (θ, θ0 ) − s∞ (θ0 , θ0 ) < 0, ∀θ 6= θ0 , a.s.

Suppose that θ∗ is a limit point of θˆn (any sequence from a compact set has at least one limit point). Since θˆ n is a maximizer, independent of n, we must have

4.3. THE SCORE FUNCTION

60

s∞ (θ∗ , θ0 ) − s∞ (θ0 , θ0 ) ≥ 0. These last two inequalities imply that θ∗ = θ0 , a.s. Thus there is only one limit point, and it is equal to the true parameter value with probability one. In other words, lim θˆ = θ0 , a.s.

n→∞

This completes the proof of strong consistency of the MLE. One can use weaker assumptions to prove weak consistency (convergence in probability to θ 0 ) of the MLE. This is omitted here. Note that almost sure convergence implies convergence in probability.

4.3. The score function Differentiability: Assume that sn (θ) is twice continuously differentiable in a neighborhood N(θ0 ) of θ0 , at least when n is large enough. To maximize the log-likelihood function, take derivatives: gn (Y, θ) = Dθ sn (θ) =

1 n ∑ Dθ ln f (yt |xx, θ) n t=1

1 n ≡ ∑ gt (θ). n t=1

4.3. THE SCORE FUNCTION

61

This is the score vector (with dim K × 1). Note that the score function has Y as an argument, which implies that it is a random function. Y (and any exogeneous variables) will often be suppressed for clarity, but one should not forget that they are still there. The ML estimator θˆ sets the derivatives to zero: n ˆ = 1 ∑ gt (θ) ˆ ≡ 0. gn (θ) n t=1

We will show that Eθ [gt (θ)] = 0, ∀t. This is the expectation taken with respect to the density f (θ), not necessarily f (θ0 ) . Eθ [gt (θ)] =

Z

=

Z

=

Z

[Dθ ln f (yt |xt , θ)] f (yt |x, θ)dyt 1 [Dθ f (yt |xt , θ)] f (yt |xt , θ)dyt f (yt |xt , θ) Dθ f (yt |xt , θ)dyt .

Given some regularity conditions on boundedness of Dθ f , we can switch the order of integration and differentiation, by the dominated convergence theorem. This gives Eθ [gt (θ)] = Dθ

Z

f (yt |xt , θ)dyt

= Dθ 1 = 0 where we use the fact that the integral of the density is 1.

• So Eθ (gt (θ) = 0 : the expectation of the score vector is zero.

• This hold for all t, so it implies that Eθ gn (Y, θ) = 0.

4.4. ASYMPTOTIC NORMALITY OF MLE

62

4.4. Asymptotic normality of MLE Recall that we assume that sn (θ) is twice continuously differentiable. Take a first ˆ about the true value θ0 : order Taylor’s series expansion of g(Y, θ) ˆ = g(θ0 ) + (Dθ0 g(θ∗ )) θˆ − θ0 0 ≡ g(θ) or with appropriate definitions

H(θ∗ ) θˆ − θ0 = −g(θ0 ),

where θ∗ = λθˆ + (1 − λ)θ0 , 0 < λ < 1. Assume H(θ∗ ) is invertible (we’ll justify this in a minute). So √ √ n θˆ − θ0 = −H(θ∗ )−1 ng(θ0 )

Now consider H(θ∗ ). This is

H(θ∗ ) = Dθ0 g(θ∗ ) = D2θ sn (θ∗ ) =

1 n 2 ∑ Dθ ln ft (θ∗) n t=1

where the notation D2θ sn (θ) ≡

∂2 sn (θ) . ∂θ∂θ0

Given that this is an average of terms, it should usually be the case that this satisfies a strong law of large numbers (SLLN). Regularity conditions are a set of assumptions that guarantee that this will happen. There are different sets of assumptions that can be used to justify appeal to different SLLN’s. For example, the D2θ ln ft (θ∗ ) must not be too strongly dependent over time, and their variances must not become infinite. We


63

don’t assume any particular set here, since the appropriate assumptions will depend upon the particularities of a given model. However, we assume that a SLLN applies. Also, since we know that θˆ is consistent, and since θ∗ = λθˆ + (1 − λ)θ0 , we have a.s.

that θ∗ → θ0 . Also, by the above differentiability assumtion, H(θ) is continuous in θ. Given this, H(θ∗ ) converges to the limit of it’s expectation: a.s. H(θ∗ ) → lim E D2θ sn (θ0 ) = H∞ (θ0 ) < ∞ n→∞

This matrix converges to a finite limit.

Re-arranging orders of limits and differentiation, which is legitimate given regularity conditions, we get H∞ (θ0 ) = D2θ lim E (sn (θ0 )) n→∞

= D2θ s∞ (θ0 , θ0 ) We’ve already seen that s∞ (θ, θ0 ) < s∞ (θ0 , θ0 ) i.e., θ0 maximizes the limiting objective function. Since there is a unique maximizer, and by the assumption that sn (θ) is twice continuously differentiable (which holds in the limit), then H∞ (θ0 ) must be negative definite, and therefore of full rank. Therefore the previous inversion is justified, asymptotically, and we have (4.4.1)

√

a.s. √ n θˆ − θ0 → −H∞ (θ0 )−1 ng(θ0 ).


Now consider

√

64

ng(θ0 ). This is √ √ ngn (θ0 ) = nDθ sn (θ) √ n n = ∑ Dθ ln ft (yt |xt , θ0) n t=1 1 n = √ ∑ gt (θ0 ) n t=1

We’ve already seen that Eθ [gt (θ)] = 0. As such, it is reasonable to assume that a CLT applies. a.s.

Note that gn (θ0 ) → 0, by consistency. To avoid this collapse to a degenerate r.v. (a √ constant vector) we need to scale by n. A generic CLT states that, for Xn a random vector that satisfies certain conditions, d

Xn − E(Xn ) → N(0, limV (Xn )) The “certain conditions” that Xn must satisfy depend on the case at hand. Usually, Xn √ will be of the form of an average, scaled by n: Xn = This is the case for

√

n √ ∑t=1 Xt n n

ng(θ0 ) for example. Then the properties of Xn depend on the

properties of the Xt . For example, if the Xt have finite variances and are not too strongly dependent, then a CLT for dependent processes will apply. Supposing that a CLT √ applies, and noting that E( ngn (θ0 ) = 0, we get √

d

I∞ (θ0 )−1/2 ngn (θ0 ) → N [0, IK ]


65

where I∞ (θ0 ) =

=

lim Eθ0 n [gn (θ0 )] [gn (θ0 )]0

n→∞

lim Vθ0

n→∞

√

ngn (θ0 )

This can also be written as √

(4.4.2)

d

ngn (θ0 ) → N [0, I∞ (θ0 )]

• I∞ (θ0 ) is known as the information matrix. • Combining [4.4.1] and [4.4.2], we get √

a n θˆ − θ0 ∼ N 0, H∞ (θ0 )−1 I∞ (θ0 )H∞ (θ0 )−1 .

The MLE estimator is asymptotically normally distributed. √ D EFINITION 1 (CAN). An estimator θˆ of a parameter θ0 is n-consistent and asymptotically normally distributed if (4.4.3)

√

d n θˆ − θ0 → N (0,V∞ )

where V∞ is a finite positive definite matrix.

p √ There do exist, in special cases, estimators that are consistent such that n θˆ − θ0 → √ 0. These are known as superconsistent estimators, since normally, n is the highest factor that we can multiply by an still get convergence to a stable limiting distribution. D EFINITION 2 (Asymptotic unbiasedness). An estimator θˆ of a parameter θ0 is asymptotically unbiased if (4.4.4)

ˆ = θ. lim Eθ (θ)

n→∞

4.6. THE INFORMATION MATRIX EQUALITY

66

Estimators that are CAN are asymptotically unbiased, though not all consistent estimators are asymptotically unbiased. Such cases are unusual, though. An example is

E XERCISE 4.5. Consider an estimator θˆ with density ˆ = f (θ)

1 − n1 , θˆ = θ0 1 ˆ n, θ =

n

Show that this estimator is consistent but asymptotically biased. Also ask yourself how you could define an estimator that would have this density.

4.6. The information matrix equality We will show that H∞ (θ) = −I∞ (θ). Let ft (θ) be short for f (yt |xt , θ) 1 =

Z

ft (θ)dy, so

0 =

Z

Dθ ft (θ)dy

=

Z

(Dθ ln ft (θ)) ft (θ)dy

Now differentiate again: Z Z 2 Dθ ln ft (θ) ft (θ)dy + [Dθ ln ft (θ)] Dθ0 ft (θ)dy 0 = 2 Z = Eθ Dθ ln ft (θ) + [Dθ ln ft (θ)] [Dθ0 ln ft (θ)] ft (θ)dy

(4.6.1)

= Eθ D2θ ln ft (θ) + Eθ [Dθ ln ft (θ)] [Dθ0 ln ft (θ)] = Eθ [Ht (θ)] + Eθ [gt (θ)] [gt (θ)]0

4.6. THE INFORMATION MATRIX EQUALITY

Now sum over n and multiply by Eθ

n

67

1 n

"

n

1 1 [Ht (θ)] = −Eθ [gt (θ)] [gt (θ)]0 ∑ ∑ n t=1 n t=1

#

The scores gt and gs are uncorrelated for t 6= s, since for t > s, ft (yt |y1 , ..., yt−1 , θ) has conditioned on prior information, so what was random in s is fixed in t. (This forms the basis for a specification test proposed by White: if the scores appear to be correlated one may question the specification of the model). This allows us to write Eθ [H(θ)] = −Eθ n [g(θ)][g(θ)]0

since all cross products between different periods expect to zero. Finally take limits, we get (4.6.2)

H∞ (θ) = −I∞ (θ).

This holds for all θ, in particular, for θ0 . Using this, √ simplifies to (4.6.3)

a.s. n θˆ − θ0 → N 0, H∞ (θ0 )−1 I∞ (θ0 )H∞ (θ0 )−1 a.s. √ n θˆ − θ0 → N 0, I∞ (θ0 )−1

To estimate the asymptotic variance, we need estimators of H∞ (θ0 ) and I∞ (θ0 ). We can use n

ˆ ˆ 0 I\ ∞ (θ0 ) = n ∑ gt (θ)gt (θ) t=1

ˆ H\ ∞ (θ0 ) = H(θ).

4.7. THE CRAMÉR-RAO LOWER BOUND

68

Note, one can’t use ˆ ˆ 0 I\ ∞ (θ0 ) = n gn (θ) gn (θ)

to estimate the information matrix. Why not?

From this we see that there are alternative ways to estimate V∞ (θ0 ) that are all valid. These include \ V\ ∞ (θ0 ) = −H∞ (θ0 )

−1

−1

\ V\ ∞ (θ0 ) = I∞ (θ0 )

−1

−1

\ \ \ V\ ∞ (θ0 ) = H∞ (θ0 ) I∞ (θ0 )H∞ (θ0 )

These are known as the inverse Hessian, outer product of the gradient (OPG) and sandwich estimators, respectively. The sandwich form is the most robust, since it coincides with the covariance estimator of the quasi-ML estimator.

4.7. The Cramér-Rao lower bound T HEOREM 3. [Cramer-Rao Lower Bound] The limiting variance of a CAN estima˜ minus the inverse of the information matrix is a positive semidefinite tor of θ0 , say θ, matrix. Proof: Since the estimator is CAN, it is asymptotically unbiased, so lim Eθ (θ˜ − θ) = 0

n→∞

Differentiate wrt θ0 : Dθ0 lim Eθ (θ˜ − θ) = n→∞

lim

n→∞

Z

Dθ0 f (Y, θ) θ˜ − θ dy

= 0 (this is a K × K matrix of zeros).


69

Noting that Dθ0 f (Y, θ) = f (θ)Dθ0 ln f (θ), we can write lim

n→∞

Z

Z ˜θ − θ f (θ)Dθ0 ln f (θ)dy + lim f (Y, θ)Dθ0 θ˜ − θ dy = 0. n→∞

R Now note that Dθ0 θ˜ − θ = −IK , and f (Y, θ)(−IK )dy = −IK . With this we have lim

n→∞

Z

θ˜ − θ f (θ)Dθ0 ln f (θ)dy = IK .

Playing with powers of n we get lim

n→∞

Z

√

n θ˜ − θ

√ 1 n [Dθ0 ln f (θ)] f (θ)dy = IK {z } |n

Note that the bracketed part is just the transpose of the score vector, g(θ), so we can write lim Eθ

n→∞

√

n θ˜ − θ

√ ng(θ)0 = IK

√ ˜ n θ − θ , for θ˜ any CAN √ estimator, is an identity matrix. Using this, suppose the variance of n θ˜ − θ tends This means that the covariance of the score function with

˜ Therefore, to V∞ (θ).

  √   ˜ ˜ n θ−θ V (θ) IK = ∞ . V∞  √ ng(θ) IK I∞ (θ)

(4.7.1)

Since this is a covariance matrix, it is positive semi-definite. Therefore, for any K -vector α, h

i



α0 −α0 I∞−1 (θ) 

This simplifies to

˜ V∞ (θ) IK

IK I∞ (θ)

 

α −I∞

˜ − I∞−1 (θ) α ≥ 0. α0 V∞ (θ)

(θ)−1 α



 ≥ 0.


70

˜ − I∞ (θ) is positive semidefinite. This conludes the proof. Since α is arbitrary, V∞ (θ) This means that I∞−1 (θ) is a lower bound for the asymptotic variance of a CAN estimator. D EFINITION 4.7.1. (Asymptotic efficiency) Given two CAN estimators of a paramˆ θˆ is asymptotically efficient with respect to θ˜ if V∞ (θ) ˜ −V∞ (θ) ˆ is eter θ0 , say θ˜ and θ, a positive semidefinite matrix. A direct proof of asymptotic efficiency of an estimator is infeasible, but if one can show that the asymptotic variance is equal to the inverse of the information matrix, then the estimator is asymptotically efficient. In particular, the MLE is asymptotically efficient. Summary of MLE • Consistent • Asymptotically normal (CAN) • Asymptotically efficient • Asymptotically unbiased • This is for general MLE: we haven’t specified the distribution or the linearity/nonlinearity of the estimator

EXERCISES

71

Exercises (1) Consider coin tossing with a single possibly biased coin. The density function for the random variable y = 1(heads) is y

fY (y, p0 ) = p0 (1 − p0 )1−y , y ∈ {0, 1} = 0, y ∈ / {0, 1} Suppose that we have a sample of size n. We know from above that the ML estimator is pb0 = y. ¯ We also know from the theory above that

√ a n (y¯ − p0 ) ∼ N 0, H∞ (p0 )−1 I∞ (p0 )H∞ (p0 )−1

a) find the analytical expressions for H∞ (p0 ) and I∞ (p0 ) for this problem √ b) Write an Octave program that does a Monte Carlo study that shows that n (y¯ − p0 ) is approximately normally distributed when n is large. Please give me histograms √ that show the sampling frequency of n (y¯ − p0 ) for several values of n.

(2) Consider the model yt = xt0 β + αεt where the errors follow the Cauchy (Student-t with 1 degree of freedom) density. So f (εt ) =

1 , −∞ < εt < ∞ π 1 + εt2

The Cauchy density has a shape similar to a normal density, but with much thicker tails. Thus, extremely small and large errors occur much more frequently with this density than would happen if the errors were normally distributed. Find the score 0 function gn (θ) where θ = β0 α .

(3) Consider the model classical linear regression model yt = xt0 β + εt where εt ∼ 0 IIN(0, σ2). Find the score function gn (θ) where θ = β0 σ .

EXERCISES

72

(4) Compare the first order conditional that define the ML estimators of problems 2 and 3 and interpret the differences. Why are the first order conditions that define an efficient estimator different in the two cases?

CHAPTER 5

Asymptotic properties of the least squares estimator The OLS estimator under the classical assumptions is unbiased and BLUE, for all sample sizes. Now let’s see what happens when the sample size tends to infinity.

5.1. Consistency

βˆ = (X 0 X )−1 X 0 y = (X 0 X )−1 X 0 (X β + ε) = β0 + (X 0 X )−1 X 0 ε 0 −1 0 XX Xε = β0 + n n Consider the last two terms. By assumption limn→∞

X 0X n

= QX ⇒ limn→∞

X 0X n

−1

=

Q−1 X , since the inverse of a nonsingular matrix is a continuous function of the elements of the matrix. Considering

X 0ε n ,

X 0ε 1 n = ∑ xt εt n n t=1 Each xt εt has expectation zero, so E

X 0ε n

73

=0

5.2. ASYMPTOTIC NORMALITY

74

The variance of each term is V (xt εt ) = xt xt0 σ2 . As long as these are finite, and given a technical condition1, the Kolmogorov SLLN applies, so

1 n a.s. ∑ xt εt → 0. n t=1

This implies that a.s. βˆ → β0 .

This is the property of strong consistency: the estimator converges in almost surely to the true value. • The consistency proof does not use the normality assumption. • Remember that almost sure convergence implies convergence in probability. 5.2. Asymptotic normality We’ve seen that the OLS estimator is normally distributed under the assumption of normal errors. If the error distribution is unknown, we of course don’t know the distribution of the estimator. However, we can get asymptotic results. Assuming the distribution of ε is unknown, but the the other classical assumptions hold:

βˆ = β0 + (X 0 X )−1 X 0 ε βˆ − β0 = (X 0 X )−1 X 0 ε 0 −1 0 √ ˆ XX Xε √ n β − β0 = n n

1For application of LLN’s and CLT’s, of which there are very many to choose from, I’m going to avoid

the technicalities. Basically, as long as terms of an average have finite variances and are not too strongly dependent, one will be able to find a LLN or CLT to apply.

5.3. ASYMPTOTIC EFFICIENCY

• Now as before, • Considering

0 X √ε, n

X 0X n

−1

75

→ Q−1 X .

the limit of the variance is

lim V

n→∞

X 0ε √ n

= lim E n→∞

= σ20 QX

X 0 εε0 X n

The mean is of course zero. To get asymptotic normality, we need to apply a CLT. We assume one (for instance, the Lindeberg-Feller CLT) holds, so

Therefore,

X 0ε d √ → N 0, σ20 QX n √ ˆ d 2 −1 n β − β0 → N 0, σ0 QX

• In summary, the OLS estimator is normally distributed in small and large

samples if ε is normally distributed. If ε is not normally distributed, βˆ is asymptotically normally distributed when a CLT can be applied.

5.3. Asymptotic efficiency The least squares objective function is

s(β) =

n

∑

t=1

yt − xt0 β

2

Supposing that ε is normally distributed, the model is y = X β0 + ε,

5.3. ASYMPTOTIC EFFICIENCY

ε ∼ N(0, σ20 In ), so n

ε2 exp − t 2 f (ε) = ∏ √ 2 2σ t=1 2πσ 1

76

The joint density for y can be constructed using a change of variables. We have ε = y − X β, so

∂ε ∂y0

∂ε = In and | ∂y 0 | = 1, so

(yt − xt0 β)2 exp − f (y) = ∏ √ . 2 2σ2 t=1 2πσ n

Taking logs,

1

(yt − xt0 β)2 ln L(β, σ) = −n ln 2π − n ln σ − ∑ . 2σ2 t=1 √

n

It’s clear that the fonc for the MLE of β0 are the same as the fonc for OLS (up to multiplication by a constant), so the estimators are the same, under the present assumptions. Therefore, their properties are the same. In particular, under the classical assumptions with normality, the OLS estimator βˆ is asymptotically efficient. As we’ll see later, it will be possible to use (iterated) linear estimation methods and still achieve asymptotic efficiency even if the assumption that Var(ε) 6= σ 2 In , as long as ε is still normally distributed. This is not the case if ε is nonnormal. In general with

nonnormal errors it will be necessary to use nonlinear estimation methods to achieve asymptotically efficient estimation.

CHAPTER 6

Restrictions and hypothesis tests 6.1. Exact linear restrictions In many cases, economic theory suggests restrictions on the parameters of a model. For example, a demand function is supposed to be homogeneous of degree zero in prices and income. If we have a Cobb-Douglas (log-linear) model, ln q = β0 + β1 ln p1 + β2 ln p2 + β3 ln m + ε, then we need that k0 ln q = β0 + β1 ln kp1 + β2 ln kp2 + β3 ln km + ε, so β1 ln p1 + β2 ln p2 + β3 ln m = β1 ln kp1 + β2 ln kp2 + β3 ln km = (ln k) (β1 + β2 + β3 ) + β1 ln p1 + β2 ln p2 + β3 ln m. The only way to guarantee this for arbitrary k is to set β1 + β2 + β3 = 0, which is a parameter restriction. In particular, this is a linear equality restriction, which is probably the most commonly encountered case.

77

6.1. EXACT LINEAR RESTRICTIONS

78

6.1.1. Imposition. The general formulation of linear equality restrictions is the model y = Xβ + ε Rβ = r where R is a Q × K matrix, Q < K and r is a Q × 1 vector of constants. • We assume R is of rank Q, so that there are no redundant restrictions. • We also assume that ∃β that satisfies the restrictions: they aren’t infeasible. Let’s consider how to estimate β subject to the restrictions Rβ = r. The most obvious approach is to set up the Lagrangean 1 (y − X β)0 (y − X β) + 2λ0 (Rβ − r). n

min s(β) = β

The Lagrange multipliers are scaled by 2, which makes things less messy. The fonc are ˆ λ) ˆ = −2X 0 y + 2X 0 X βˆ R + 2R0 λ ˆ ≡0 Dβ s(β, ˆ λ) ˆ = Rβˆ R − r ≡ 0, Dλ s(β, which can be written as  We get

  

X 0X R

βˆ R ˆ λ



R0 0 

=

 

βˆ R ˆ λ



=

X 0 X R0 R

For the masochists: Stepwise Inversion



0

X 0y

−1  



r



.

X 0y r



.


Note that  

(X 0 X )−1

0

−R (X 0 X )−1

IQ

 

X 0 X R0 R

0

79



 ≡ AB 

=  

≡ 

IK

(X 0 X )−1 R0

0

−R (X 0 X )−1 R0

IK

(X 0 X )−1 R0

0

−P

≡ C,



 



and  

IK

(X 0 X )−1 R0 P−1

0

−P−1

 

IK

(X 0 X )−1 R0

0

−P



 ≡ DC

= IK+Q ,

so DAB = IK+Q DA = B−1    −1 0 0 −1 0 −1 (X X ) 0 IK (X X ) R P   B−1 =  −1 −1 0 0 −P −R (X X ) IQ   (X 0 X )−1 − (X 0 X )−1 R0 P−1 R (X 0 X )−1 (X 0X )−1 R0 P−1 , =  −1 −1 0 −1 P R (X X ) −P


80

so (everyone should start paying attention again)  

βˆ R ˆ λ





 =  

 = 



= 

(X 0 X )−1 − (X 0 X )−1 R0 P−1 R (X 0 X )−1 P−1 R (X 0 X )−1

(X 0 X )−1 R0 P−1 −P−1

 ˆβ − (X 0 X )−1 R0 P−1 Rβˆ − r   P−1 Rβˆ − r    IK − (X 0 X )−1 R0 P−1 R (X 0X )−1 R0 P−1 r  βˆ +   −1 −1 P R −P r

 

X 0y r

 

ˆ are linear functions of βˆ makes it easy to determine their disThe fact that βˆ R and λ tributions, since the distribution of βˆ is already known. Recall that for x a random vector, and for A and b a matrix and vector of constants, respectively, Var (Ax + b) = AVar(x)A0. Though this is the obvious way to go about finding the restricted estimator, an easier way, if the number of restrictions is small, is to impose them by substitution. Write

h

R1 R2

i

 

β1 β2

y = X1 β1 + X2 β2 + ε   = r

where R1 is Q × Q nonsingular. Supposing the Q restrictions are linearly independent, one can always make R1 nonsingular by reorganizing the columns of X . Then −1 β1 = R−1 1 r − R 1 R 2 β2 .


81

Substitute this into the model −1 y = X1 R−1 1 r − X1 R1 R2 β2 + X2 β2 + ε i h −1 −1 y − X1 R1 r = X2 − X1 R1 R2 β2 + ε

or with the appropriate definitions, yR = XR β2 + ε. This model satisfies the classical assumptions, supposing the restriction is true. One can estimate by OLS. The variance of βˆ 2 is as before V (βˆ 2 ) = XR0 XR

−1

σ20

Vˆ (βˆ 2 ) = XR0 XR

−1

σˆ 2

and the estimator is

where one estimates σ20 in the normal way, using the restricted model, i.e., c2 = σ 0

yR − XR βb2

0

yR − XR βb2

n − (K − Q)

To recover βˆ 1 , use the restriction. To find the variance of βˆ 1 , use the fact that it is a linear function of βˆ 2 , so 0 ˆ 2 )R0 R−1 V (βˆ 1 ) = R−1 R V ( β 2 2 1 1 −1 0 −1 0 2 0 R X X R2 R1 = R−1 σ0 2 2 2 1


82

6.1.2. Properties of the restricted estimator. We have that ˆβR = βˆ − (X 0 X )−1 R0 P−1 Rβˆ − r

= βˆ + (X 0 X )−1 R0 P−1 r − (X 0X )−1 R0 P−1 R(X 0X )−1 X 0 y = β + (X 0X )−1 X 0ε + (X 0 X )−1 R0 P−1 [r − Rβ] − (X 0 X )−1 R0 P−1 R(X 0 X )−1 X 0 ε

βˆ R − β = (X 0X )−1 X 0 ε + (X 0X )−1 R0 P−1 [r − Rβ] − (X 0X )−1 R0 P−1 R(X 0X )−1 X 0 ε Mean squared error is MSE(βˆ R ) = E (βˆ R − β)(βˆ R − β)0 Noting that the crosses between the second term and the other terms expect to zero, and that the cross of the first and third has a cancellation with the square of the third, we obtain MSE(βˆ R ) = (X 0 X )−1 σ2 + (X 0 X )−1 R0 P−1 [r − Rβ] [r − Rβ]0 P−1 R(X 0 X )−1 − (X 0 X )−1 R0 P−1 R(X 0X )−1 σ2 So, the first term is the OLS covariance. The second term is PSD, and the third term is NSD. • If the restriction is true, the second term is 0, so we are better off. True restrictions improve efficiency of estimation. • If the restriction is false, we may be better or worse off, in terms of MSE, depending on the magnitudes of r − Rβ and σ2 .

6.2. TESTING

83

6.2. Testing In many cases, one wishes to test economic theories. If theory suggests parameter restrictions, as in the above homogeneity example, one can test theory by testing parameter restrictions. A number of tests are available. 6.2.1. t-test. Suppose one has the model

y = Xβ + ε and one wishes to test the single restriction H0 :Rβ = r vs. HA :Rβ 6= r . Under H0 , with normality of the errors, Rβˆ − r ∼ N 0, R(X 0X )−1 R0 σ20 so

Rβˆ − r Rβˆ − r q = p ∼ N (0, 1) . 0 −1 0 R(X 0 X )−1 R0 σ20 σ0 R(X X ) R

c2 in place The problem is that σ20 is unknown. One could use the consistent estimator σ 0

of σ20 , but the test would only be valid asymptotically in this case. P ROPOSITION 4. (6.2.1)

N(0, 1) q 2 ∼ t(q) χ (q) q

as long as the N(0, 1) and the χ2 (q) are independent. We need a few results on the χ2 distribution. P ROPOSITION 5. If x ∼ N(µ, In ) is a vector of n independent r.v.’s., then (6.2.2)

x0 x ∼ χ2 (n, λ)

6.2. TESTING

84

where λ = ∑i µ2i = µ0 µ is the noncentrality parameter. When a χ2 r.v. has the noncentrality parameter equal to zero, it is referred to as a central χ2 r.v., and it’s distribution is written as χ2 (n), suppressing the noncentrality parameter. P ROPOSITION 6. If the n dimensional random vector x ∼ N(0,V ), then x 0V −1 x ∼

χ2 (n).

We’ll prove this one as an indication of how the following unproven propositions could be proved. Proof: Factor V −1 as PP0 (this is the Cholesky factorization). Then consider y = P0 x. We have y ∼ N(0, P0V P) but V PP0 = In P0V PP0 = P0 so PV P0 = In and thus y ∼ N(0, In ). Thus y0 y ∼ χ2 (n) but y0 y = x0 PP0 x = xV −1 x and we get the result we wanted. A more general proposition which implies this result is P ROPOSITION 7. If the n dimensional random vector x ∼ N(0,V ), then (6.2.3) if and only if BV is idempotent.

x0 Bx ∼ χ2 (ρ(B))

6.2. TESTING

85

An immediate consequence is

P ROPOSITION 8. If the random vector (of dimension n) x ∼ N(0, I), and B is idempotent with rank r, then x0 Bx ∼ χ2 (r).

(6.2.4) Consider the random variable

ε 0 MX ε εˆ 0 εˆ = σ20 σ20 0 ε ε MX = σ0 σ0 ∼ χ2 (n − K)

P ROPOSITION 9. If the random vector (of dimension n) x ∼ N(0, I), then Ax and x0 Bx are independent if AB = 0.

Now consider (remember that we have only one restriction in this case) ˆ

√ Rβ−r σ0 R(X 0 X)−1 R0 q 0 = εˆ εˆ (n−K)σ20

Rβˆ − r p c0 R(X 0 X )−1 R0 σ

This will have the t(n − K) distribution if βˆ and εˆ 0 εˆ are independent. But βˆ = β +

(X 0 X )−1 X 0 ε and

(X 0 X )−1 X 0 MX = 0, so

Rβˆ − r Rβˆ − r p ∼ t(n − K) = σˆ Rβˆ c0 R(X 0X )−1 R0 σ

6.2. TESTING

86

In particular, for the commonly encountered test of significance of an individual coefficient, for which H0 : βi = 0 vs. H0 : βi 6= 0 , the test statistic is βˆ i ∼ t(n − K) σˆ βi ˆ • Note: the t− test is strictly valid only if the errors are actually normally distributed. If one has nonnormal errors, one could use the above asymptotic result to justify taking critical values from the N(0, 1) distribution, since t(n − d

K) → N(0, 1) as n → ∞. In practice, a conservative procedure is to take critical

values from the t distribution if nonnormality is suspected. This will reject H0 less often since the t distribution is fatter-tailed than is the normal.

6.2.2. F test. The F test allows testing multiple restrictions jointly. P ROPOSITION 10. If x ∼ χ2 (r) and y ∼ χ2 (s), then x/r ∼ F(r, s) y/s

(6.2.5)

provided that x and y are independent. P ROPOSITION 11. If the random vector (of dimension n) x ∼ N(0, I), then x 0 Ax and x0 Bx are independent if AB = 0. Using these results, and previous results on the χ2 distribution, it is simple to show that the following statistic has the F distribution:

F=

Rβˆ − r

0

R (X 0 X )−1 R0 qσˆ 2

A numerically equivalent expression is

−1

Rβˆ − r

∼ F(q, n − K).

6.2. TESTING

87

(ESSR − ESSU ) /q ∼ F(q, n − K). ESSU /(n − K)

• Note: The F test is strictly valid only if the errors are truly normally distributed. The following tests will be appropriate when one cannot assume normally distributed errors. 6.2.3. Wald-type tests. The Wald principle is based on the idea that if a restriction is true, the unrestricted model should “approximately” satisfy the restriction. Given that the least squares estimator is asymptotically normally distributed: √ ˆ d n β − β0 → N 0, σ20 Q−1 X then under H0 : Rβ0 = r, we have √ ˆ d 2 −1 0 n Rβ − r → N 0, σ0 RQX R so by Proposition [6]

n Rβˆ − r

0

0 σ20 RQ−1 X R

−1

d ˆ Rβ − r → χ2 (q)

2 Note that Q−1 X or σ0 are not observable. The test statistic we use substitutes the con-

sistent estimators. Use (X 0 X /n)−1 as the consistent estimator of Q−1 X . With this, there is a cancellation of n0 s, and the statistic to use is

Rβˆ − r

0

c2 R(X 0 X )−1 R0 σ 0

−1

d Rβˆ − r → χ2 (q)

• The Wald test is a simple way to test restrictions without having to estimate the restricted model. • Note that this formula is similar to one of the formulae provided for the F test.

6.2. TESTING

88

6.2.4. Score-type tests (Rao tests, Lagrange multiplier tests). In some cases, an unrestricted model may be nonlinear in the parameters, but the model is linear in the parameters under the null hypothesis. For example, the model y = (X β)γ + ε is nonlinear in β and γ, but is linear in β under H0 : γ = 1. Estimation of nonlinear models is a bit more complicated, so one might prefer to have a test based upon the restricted, linear model. The score test is useful in this situation. • Score-type tests are based upon the general principle that the gradient vector of the unrestricted model, evaluated at the restricted estimate, should be asymptotically normally distributed with mean zero, if the restrictions are true. The original development was for ML estimation, but the principle is valid for a wide variety of estimation methods. We have seen that −1 Rβˆ − r R(X 0X )−1 R0 −1 ˆ = P Rβ − r

ˆ = λ

Given that

√ ˆ d −1 0 2 n Rβ − r → N 0, σ0 RQX R

under the null hypothesis,

√ ˆ d 2 −1 −1 0 −1 nλ → N 0, σ0 P RQX R P or

√ ˆ d 0 −1 nλ → N 0, σ20 lim n (nP)−1 RQ−1 R P X

6.2. TESTING

89

since the n’s cancel and inserting the limit of a matrix of constants changes nothing. However, lim nP = lim nR(X 0X )−1 R0 0 −1 XX = lim R R0 n 0 = RQ−1 X R

So there is a cancellation and we get

In this case,

√ ˆ d nλ → N 0, σ20 lim nP−1 0 −1 0 d 2 ˆλ0 R(X X ) R λ ˆ → χ (q) 2 σ0

since the powers of n cancel. To get a usable test statistic substitute a consistent estimator of σ20 .

• This makes it clear why the test is sometimes referred to as a Lagrange multiplier test. It may seem that one needs the actual Lagrange multipliers to calculate this. If we impose the restrictions by substitution, these are not available. Note that the test can be written as 0 ˆ ˆ (X 0X )−1 R0 λ R0 λ d → χ2 (q) 2 σ0 However, we can use the fonc for the restricted estimator: ˆ −X 0 y + X 0 X βˆ R + R0 λ

6.2. TESTING

90

to get that ˆ = X 0 (y − X βˆ R) R0 λ = X 0 εˆ R Substituting this into the above, we get εˆ 0R X (X 0X )−1 X 0 εˆ R d 2 → χ (q) σ20 but this is simply εˆ 0R

PX d εˆ R → χ2 (q). 2 σ0

To see why the test is also known as a score test, note that the fonc for restricted least squares ˆ −X 0 y + X 0 X βˆ R + R0 λ give us ˆ = X 0 y − X 0 X βˆ R R0 λ and the rhs is simply the gradient (score) of the unrestricted model, evaluated at the restricted estimator. The scores evaluated at the unrestricted estimate are identically zero. The logic behind the score test is that the scores evaluated at the restricted estimate should be approximately zero, if the restriction is true. The test is also known as a Rao test, since P. Rao first proposed it in 1948. 6.2.5. Likelihood ratio-type tests. The Wald test can be calculated using the unrestricted model. The score test can be calculated using only the restricted model. The likelihood ratio test, on the other hand, uses both the restricted and the unrestricted estimators. The test statistic is ˆ − ln L(θ) ˜ LR = 2 ln L(θ)

6.2. TESTING

91

where θˆ is the unrestricted estimate and θ˜ is the restricted estimate. To show that it is ˜ about θˆ : asymptotically χ2 , take a second order Taylor’s series expansion of ln L(θ) ˜ ' ln L(θ) ˆ + n θ˜ − θˆ 0 H(θ) ˆ θ˜ − θˆ ln L(θ) 2

ˆ ≡ 0 by the fonc and we need to (note, the first order term drops out since Dθ ln L(θ) multiply the second-order term by n since H(θ) is defined in terms of n1 ln L(θ)) so 0 ˆ θ˜ − θˆ LR ' −n θ˜ − θˆ H(θ)

ˆ → H∞ (θ0 ) = −I (θ0 ), by the information matrix equality. So As n → ∞, H(θ) 0 a LR = n θ˜ − θˆ I∞ (θ0 ) θ˜ − θˆ

We also have that, from [??] that √

a n θˆ − θ0 = I∞ (θ0 )−1 n1/2 g(θ0 ).

An analogous result for the restricted estimator is (this is unproven here, to prove this set up the Lagrangean for MLE subject to Rβ = r, and manipulate the first order conditions) : a −1 √ n θ˜ − θ0 = I∞ (θ0 )−1 In − R0 RI∞ (θ0 )−1 R0 RI∞ (θ0 )−1 n1/2 g(θ0 ).

Combining the last two equations

a −1 √ RI∞ (θ0 )−1 g(θ0 ) n θ˜ − θˆ = −n1/2 I∞ (θ0 )−1 R0 RI∞ (θ0 )−1 R0

so, substituting into [??]

h i i −1 h a LR = n1/2 g(θ0 )0 I∞ (θ0 )−1 R0 RI∞ (θ0 )−1 R0 RI∞ (θ0 )−1 n1/2 g(θ0 )

6.3. THE ASYMPTOTIC EQUIVALENCE OF THE LR, WALD AND SCORE TESTS

92

But since d

n1/2 g(θ0 ) → N (0, I∞ (θ0 )) the linear function d

RI∞ (θ0 )−1 n1/2 g(θ0 ) → N(0, RI∞ (θ0 )−1 R0 ). We can see that LR is a quadratic form of this rv, with the inverse of its variance in the middle, so d

LR → χ2 (q).

6.3. The asymptotic equivalence of the LR, Wald and score tests We have seen that the three tests all converge to χ2 random variables. In fact, they all converge to the same χ2 rv, under the null hypothesis. We’ll show that the Wald and LR tests are asymptotically equivalent. We have seen that the Wald test is asymptotically equivalent to 0 −1 a d 2 0 ˆ −r → W = n Rβˆ − r σ20 RQ−1 R χ (q) R β X Using βˆ − β0 = (X 0 X )−1 X 0ε and Rβˆ − r = R(βˆ − β0 ) we get √ √ nR(βˆ − β0 ) = nR(X 0 X )−1 X 0 ε 0 −1 XX n−1/2 X 0 ε = R n


93

Substitute this into [??] to get a

W = n

ε

−1 0

0 X Q−1 X R

a

0 σ20 RQ−1 X R

−1

0 RQ−1 X X ε

= ε0 X (X 0X )−1 R0 σ20 R(X 0X )−1 R0 a

= a

=

ε0 A(A0 A)−1 A0 ε σ20

−1

R(X 0X )−1 X 0 ε

ε0 PR ε σ20

where PR is the projection matrix formed by the matrix X (X 0X )−1 R0 .

• Note that this matrix is idempotent and has q columns, so the projection matrix has rank q.

Now consider the likelihood ratio statistic a

LR = n1/2 g(θ0 )0 I (θ0 )−1 R0 RI (θ0 )−1 R0

−1

RI (θ0 )−1 n1/2 g(θ0 )

Under normality, we have seen that the likelihood function is √ 1 (y − X β)0 (y − X β) ln L(β, σ) = −n ln 2π − n ln σ − . 2 σ2 Using this, 1 g(β0 ) ≡ Dβ ln L(β, σ) n 0 X (y − X β0 ) = nσ2 X 0ε = nσ2


94

Also, by the information matrix equality: I (θ0 ) = −H∞ (θ0 )

= lim −Dβ0 g(β0 ) = lim −Dβ0 = lim

X 0 (y − X β0 ) nσ2

X 0X nσ2

QX σ2

= so

I (θ0 )−1 = σ2 Q−1 X

Substituting these last expressions into [??], we get a

LR = ε0 X 0 (X 0X )−1 R0 σ20 R(X 0 X )−1 R0 a

=

ε0 PR ε σ20

−1

R(X 0 X )−1 X 0 ε

a

= W This completes the proof that the Wald and LR tests are asymptotically equivalent. Similarly, one can show that, under the null hypothesis, a

a

a

qF = W = LM = LR • The proof for the statistics except for LR does not depend upon normality of the errors, as can be verified by examining the expressions for the statistics. • The LR statistic is based upon distributional assumptions, since one can’t write the likelihood function without them.


95

• However, due to the close relationship between the statistics qF and LR, supposing normality, the qF statistic can be thought of as a pseudo-LR statistic, in that it’s like a LR statistic in that it uses the value of the objective functions of the restricted and unrestricted models, but it doesn’t require distributional assumptions. • The presentation of the score and Wald tests has been done in the context of the linear model. This is readily generalizable to nonlinear models and/or other estimation methods.

Though the four statistics are asymptotically equivalent, they are numerically different in small samples. The numeric values of the tests also depend upon how σ 2 is estimated, and we’ve already seen than there are several ways to do this. For example all of the following are consistent for σ2 under H0

εˆ 0 εˆ n−k εˆ 0 εˆ n εˆ 0R εˆ R n−k+q εˆ 0R εˆ R n

and in general the denominator call be replaced with any quantity a such that lim a/n = 1. It can be shown, for linear regression models subject to linear restrictions, and if εˆ 0 εˆ n

is used to calculate the Wald test and

εˆ 0R εˆ R n

is used for the score test, that

W > LR > LM.

6.5. CONFIDENCE INTERVALS

96

For this reason, the Wald test will always reject if the LR test rejects, and in turn the LR test rejects if the LM test rejects. This is a bit problematic: there is the possibility that by careful choice of the statistic used, one can manipulate reported results to favor or disfavor a hypothesis. A conservative/honest approach would be to report all three test statistics when they are available. In the case of linear models with normal errors the F test is to be preferred, since asymptotic approximations are not an issue. The small sample behavior of the tests can be quite different. The true size (probability of rejection of the null when the null is true) of the Wald test is often dramatically higher than the nominal size associated with the asymptotic distribution. Likewise, the true size of the score test is often smaller than the nominal size.

6.4. Interpretation of test statistics Now that we have a menu of test statistics, we need to know how to use them.

6.5. Confidence intervals Confidence intervals for single coefficients are generated in the normal manner. Given the t statistic t(β) =

βˆ − β cβˆ σ

a 100 (1 − α) % confidence interval for β0 is defined by the bounds of the set of β such that t(β) does not reject H0 : β0 = β, using a α significance level: C(α) = {β : −cα/2 < The set of such β is the interval

βˆ − β < cα/2 } cβˆ σ

cβˆ cα/2 βˆ ± σ

6.6. BOOTSTRAPPING

97

A confidence ellipse for two coefficients jointly would be, analogously, the set of {β1 , β2 } such that the F (or some other test statistic) doesn’t reject at the specified critical value. This generates an ellipse, if the estimators are correlated. • The region is an ellipse, since the CI for an individual coefficient defines a (in-

finitely long) rectangle with total prob. mass 1 − α, since the other coefficient is marginalized (e.g., can take on any value). Since the ellipse is bounded in both dimensions but also contains mass 1 − α, it must extend beyond the

bounds of the individual CI. • From the pictue we can see that: – Rejection of hypotheses individually does not imply that the joint test will reject. – Joint rejection does not imply individal tests will reject.

6.6. Bootstrapping When we rely on asymptotic theory to use the normal distribution-based tests and confidence intervals, we’re often at serious risk of making important errors. If the sample size is small and errors are highly nonnormal, the small sample distribution √ of n βˆ − β0 may be very different than its large sample distribution. Also, the distributions of test statistics may not resemble their limiting distributions at all. A

means of trying to gain information on the small sample distribution of test statistics and estimators is the bootstrap. We’ll consider a simple example, just to get the main idea.

6.6. BOOTSTRAPPING

F IGURE 6.5.1. Joint and Individual Confidence Regions

98

6.6. BOOTSTRAPPING

99

Suppose that y

=

X β0 + ε

ε

∼

IID(0, σ20)

X is nonstochastic Given that the distribution of ε is unknown, the distribution of βˆ will be unknown in small samples. However, since we have random sampling, we could generate artificial data. The steps are: (1) Draw n observations from εˆ with replacement. Call this vector ε˜ j (it’s a n × 1).

(2) Then generate the data by y˜ j = X βˆ + ε˜ j (3) Now take this and estimate β˜ j = (X 0X )−1 X 0y˜ j . (4) Save β˜ j (5) Repeat steps 1-4, until we have a large number, J, of β˜ j . With this, we can use the replications to calculate the empirical distribution of β˜ j . One way to form a 100(1-α)% confidence interval for β0 would be to order the β˜ j from smallest to largest, and drop the first and last Jα/2 of the replications, and use the remaining endpoints as the limits of the CI. Note that this will not give the shortest CI if the empirical distribution is skewed. ˆ for • Suppose one was interested in the distribution of some function of β, example a test statistic. Simple: just calculate the transformation for each j, and work with the empirical distribution of the transformation.

6.7. TESTING NONLINEAR RESTRICTIONS, AND THE DELTA METHOD

100

• If the assumption of iid errors is too strong (for example if there is heteroscedasticity or autocorrelation, see below) one can work with a bootstrap defined by sampling from (y, x) with replacement. • How to choose J: J should be large enough that the results don’t change with repetition of the entire bootstrap. This is easy to check. If you find the results change a lot, increase J and try again. • The bootstrap is based fundamentally on the idea that the empirical distribution of the sample data converges to the actual sampling distribution as n becomes large, so statistics based on sampling from the empirical distribution should converge in distribution to statistics based on sampling from the actual sampling distribution. • In finite samples, this doesn’t hold. At a minimum, the bootstrap is a good way to check if asymptotic theory results offer a decent approximation to the small sample distribution. 6.7. Testing nonlinear restrictions, and the Delta Method Testing nonlinear restrictions of a linear model is not much more difficult, at least when the model is linear. Since estimation subject to nonlinear restrictions requires nonlinear estimation methods, which are beyond the score of this course, we’ll just consider the Wald test for nonlinear restrictions on a linear model. Consider the q nonlinear restrictions r(β0 ) = 0. where r(·) is a q-vector valued function. Write the derivative of the restriction evaluated at β as Dβ0 r(β) β = R(β)


101

We suppose that the restrictions are not redundant in a neighborhood of β 0 , so that ρ(R(β)) = q ˆ about β0 : in a neighborhood of β0 . Take a first order Taylor’s series expansion of r(β) ˆ = r(β0 ) + R(β∗ )(βˆ − β0 ) r(β) where β∗ is a convex combination of βˆ and β0 . Under the null hypothesis we have ˆ = R(β∗ )(βˆ − β0 ) r(β) Due to consistency of βˆ we can replace β∗ by β0 , asymptotically, so √ ˆ a √ nr(β) = nR(β0 )(βˆ − β0 ) We’ve already seen the distribution of

√

n(βˆ − β0 ). Using this we get

√ ˆ d −1 0 2 nr(β) → N 0, R(β0 )QX R(β0 ) σ0 . Considering the quadratic form −1 ˆ 0 R(β0 )Q−1 R(β0 )0 ˆ nr(β) r(β) X σ20

d

→ χ2 (q)

under the null hypothesis. Substituting consistent estimators for β0, QX and σ20 , the resulting statistic is −1 0X )−1 R(β) ˆ 0 R(β)(X ˆ ˆ 0 ˆ r(β) r(β) under the null hypothesis.

c2 σ

d

→ χ2 (q)


102

• This is known in the literature as the Delta method, or as Klein’s approximation. • Since this is a Wald test, it will tend to over-reject in finite samples. The score and LR tests are also possibilities, but they require estimation methods for nonlinear models, which aren’t in the scope of this course. Note that this also gives a convenient way to estimate nonlinear functions and associated asymptotic confidence intervals. If the nonlinear function r(β 0 ) is not hypothesized to be zero, we just have √ ˆ d −1 0 2 n r(β) − r(β0 ) → N 0, R(β0 )QX R(β0 ) σ0 so an approximation to the distribution of the function of the estimator is ˆ ≈ N(r(β0 ), R(β0 )(X 0 X )−1 R(β0 )0 σ2 ) r(β) 0 For example, the vector of elasticities of a function f (x) is η(x) =

x ∂ f (x) ∂x f (x)

where means element-by-element multiplication. Suppose we estimate a linear function y = x0 β + ε. The elasticities of y w.r.t. x are η(x) =

β x x0 β

(note that this is the entire vector of elasticities). The estimated elasticities are b(x) = η

βˆ x x0 βˆ


103

To calculate the estimated standard errors of all five elasticites, use R(β) =

=

∂η(x) ∂β0   x1 0 · · · 0  ..  0 x2 .   . ..  .. . 0   0 · · · 0 xk





        0 x β −        

β1 x21

0

0 .. .

β2 x22

0

···

(x0 β)2

··· ..

.

0

0 .. . 0 βk x2k

        

.

ˆ Note that the elasticity and the stanTo get a consistent estimator just substitute in β. dard error are functions of x. The program ExampleDeltaMethod.m shows how this can be done. In many cases, nonlinear restrictions can also involve the data, not just the parameters. For example, consider a model of expenditure shares. Let x(p, m) be a demand funcion, where p is prices and m is income. An expenditure share system for G goods is si (p, m) =

pi xi (p, m) , i = 1, 2, ..., G. m

Now demand must be positive, and we assume that expenditures sum to income, so we have the restrictions

G

0 ≤ si (p, m) ≤ 1, ∀i

∑ si(p, m)

=

1

i=1

Suppose we postulate a linear model for the expenditure shares: si (p, m) = βi1 + p0 βip + mβim + εi

6.8. EXAMPLE: THE NERLOVE DATA

104

It is fairly easy to write restrictions such that the shares sum to one, but the restriction that the shares lie in the [0, 1] interval depends on both parameters and the values of p and m. It is impossible to impose the restriction that 0 ≤ si (p, m) ≤ 1 for all possible p and m. In such cases, one might consider whether or not a linear model is a reasonable specification. 6.8. Example: the Nerlove data Remember that we in a previous example (section 3.8.3) that the OLS results for the Nerlove model are

********************************************************* OLS estimation results Observations 145 R-squared 0.925955 Sigma-squared 0.153943 Results (Ordinary var-cov estimator) constant output labor fuel capital

estimate -3.527 0.720 0.436 0.427 -0.220

st.err. 1.774 0.017 0.291 0.100 0.339

t-stat. -1.987 41.244 1.499 4.249 -0.648

p-value 0.049 0.000 0.136 0.000 0.518

********************************************************* Note that sK = βK < 0, and that βL + βF + βK 6= 1.

Remember that if we have constant returns to scale, then βQ = 1, and if there is

homogeneity of degree 1 thenβL + βF + βK = 1. We can test these hypotheses either separately or jointly. NerloveRestrictions.m imposes and tests CRTS and then HOD1. From it we obtain the results that follow:


105

Imposing and testing HOD1

******************************************************* Restricted LS estimation results Observations 145 R-squared 0.925652 Sigma-squared 0.155686

estimate

st.err.

t-stat.

p-value

-4.691

0.891

-5.263

0.000

output

0.721

0.018

41.040

0.000

labor

0.593

0.206

2.878

0.005

fuel

0.414

0.100

4.159

0.000

-0.007

0.192

-0.038

0.969

constant

capital

******************************************************* Value

p-value

F

0.574

0.450

Wald

0.594

0.441

LR

0.593

0.441

Score

0.592

0.442

Imposing and testing CRTS

*******************************************************


106

Restricted LS estimation results Observations 145 R-squared 0.790420 Sigma-squared 0.438861

estimate

st.err.

t-stat.

p-value

-7.530

2.966

-2.539

0.012

output

1.000

0.000

Inf

0.000

labor

0.020

0.489

0.040

0.968

fuel

0.715

0.167

4.289

0.000

capital

0.076

0.572

0.132

0.895

constant

******************************************************* Value

p-value

F

256.262

0.000

Wald

265.414

0.000

LR

150.863

0.000

93.771

0.000

Score

Notice that the input price coefficients in fact sum to 1 when HOD1 is imposed. HOD1 is not rejected at usual significance levels (e.g., α = 0.10). Also, R 2 does not drop much when the restriction is imposed, compared to the unrestricted results. For CRTS, you should note that βQ = 1, so the restriction is satisfied. Also note that the hypothesis that βQ = 1 is rejected by the test statistics at all reasonable significance


107

levels. Note that R2 drops quite a bit when imposing CRTS. If you look at the unrestricted estimation results, you can see that a t-test for βQ = 1 also rejects, and that a confidence interval for βQ does not overlap 1. From the point of view of neoclassical economic theory, these results are not anomalous: HOD1 is an implication of the theory, but CRTS is not.

E XERCISE 12. Modify the NerloveRestrictions.m program to impose and test the restrictions jointly.

The Chow test. Since CRTS is rejected, let’s examine the possibilities more carefully. Recall that the data is sorted by output (the third column). Define 5 subsamples of firms, with the first group being the 29 firms with the lowest output levels, then the next 29 firms, etc. The five subsamples can be indexed by j = 1, 2, ..., 5, where j = 1 for t = 1, 2, ...29, j = 2 for t = 30, 31, ...58, etc. Define a piecewise linear model (6.8.1)

j

j

j

j

j

lnCt = β1 + β2 ln Qt + β3 ln PLt + β4 ln PFt + β5 ln PKt + εt

where j is a superscript (not a power) that inicates that the coefficients may be different according to the subsample in which the observation falls. That is, the coefficients depend upon j which in turn depends upon t. Note that the first column of nerlove.data indicates this way of breaking up the sample.     y1   X1 0 · · ·     y2   0 X2     .   .  ..  =  .. (6.8.2) X3          X4       0 y5

The new model may be written as     1 1 0  β   ε   2   2   β   ε        .    +  ..               0         ε5 β5 X5


108

where y1 is 29×1, X1 is 29×5, β j is the 5×1 vector of coefficient for the jth subsample, and ε j is the 29 × 1 vector of errors for the jth subsample. The Octave program Restrictions/ChowTest.m estimates the above model. It also tests the hypothesis that the five subsamples share the same parameter vector, or in other words, that there is coefficient stability across the five subsamples. The null to test is that the parameter vectors for the separate groups are all the same, that is, β1 = β 2 = β 3 = β 4 = β 5 This type of test, that parameters are constant across different sets of data, is sometimes referred to as a Chow test. • There are 20 restrictions. If that’s not clear to you, look at the Octave program. • The restrictions are rejected at all conventional significance levels. Since the restrictions are rejected, we should probably use the unrestricted model for analysis. What is the pattern of RTS as a function of the output group (small to large)? Figure 6.8.1 plots RTS. We can see that there is increasing RTS for small firms, but that RTS is approximately constant for large firms.


109

F IGURE 6.8.1. RTS as a function of firm size 2.6

RTS

2.4 2.2 2 1.8 1.6 1.4 1.2 1 0.8

1

1.5

2

2.5

3 Output group

3.5

4

4.5

5

(1) Using the Chow test on the Nerlove model, we reject that there is coefficient stability across the 5 groups. But perhaps we could restrict the input price coefficients to be the same but let the constant and output coefficients vary by group size. This new model is (6.8.3)

j

j

lnCi = β1 + β2 ln Qi + β3 ln PLi + β4 ln PFi + β5 ln PKi + εi (a) estimate this model by OLS, giving R, estimated standard errors for coefficients, t-statistics for tests of significance, and the associated p-values. Interpret the results in detail. (b) Test the restrictions implied by this model using the F, Wald, score and likelihood ratio tests. Comment on the results.


110

(c) Plot the estimated RTS parameters as a function of firm size. Compare the plot to that given in the notes for the unrestricted model. Comment on the results. dS = (2) For the simple Nerlove model, estimated returns to scale is RT

1 . βbq

Apply

the delta method to calculate the estimated standard error for estimated RTS. Directly test H0 : RT S = 1 versus HA : RT S 6= 1 rather than testing H0 : βQ = 1 versus HA : βQ 6= 1. Comment on the results.

(3) Perform a Monte Carlo study that generates data from the model y = −2 + 1x2 + 1x3 + ε where the sample size is 30, x2 and x3 are independently uniformly distributed on [0, 1] and ε ∼ IIN(0, 1) (a) Compare the means and standard errors of the estimated coefficients using OLS and restricted OLS, imposing the restriction that β2 + β3 = 2. (b) Compare the means and standard errors of the estimated coefficients using OLS and restricted OLS, imposing the restriction that β2 + β3 = 1. (c) Discuss the results. (4) Get the Octave scripts bootstrap_example1.m , bootstrap.m , bootstrap_resample_iid.m and myols.m figure out what they do, run them, and interpret the results.

CHAPTER 7

Generalized least squares One of the assumptions we’ve made up to now is that εt ∼ IID(0, σ2), or occasionally εt ∼ IIN(0, σ2). Now we’ll investigate the consequences of nonidentically and/or dependently distributed errors. We’ll assume fixed regressors for now, relaxing this admittedly unrealistic assumption later. The model is y = Xβ + ε E (ε) = 0

V (ε) = Σ where Σ is a general symmetric positive definite matrix (we’ll write β in place of β 0 to simplify the typing of these notes). • The case where Σ is a diagonal matrix gives uncorrelated, nonidentically distributed errors. This is known as heteroscedasticity. • The case where Σ has the same number on the main diagonal but nonzero elements off the main diagonal gives identically (assuming higher moments are also the same) dependently distributed errors. This is known as autocorrelation. 111

7.1. EFFECTS OF NONSPHERICAL DISTURBANCES ON THE OLS ESTIMATOR

112

• The general case combines heteroscedasticity and autocorrelation. This is known as “nonspherical” disturbances, though why this term is used, I have no idea. Perhaps it’s because under the classical assumptions, a joint confidence region for ε would be an n− dimensional hypersphere.

7.1. Effects of nonspherical disturbances on the OLS estimator The least square estimator is βˆ = (X 0X )−1 X 0 y = β + (X 0X )−1 X 0 ε • We have unbiasedness, as before.

• The variance of βˆ is h

E (βˆ − β)(βˆ − β)0

(7.1.1)

i

= E (X 0 X )−1 X 0 εε0 X (X 0X )−1 = (X 0 X )−1 X 0 ΣX (X 0X )−1

c2 or the probability limit σ c2 Due to this, any test statistic that is based upon σ

of is invalid. In particular, the formulas for the t, F, χ2 based tests given above do not lead to statistics with these distributions.

• βˆ is still consistent, following exactly the same argument given before. • If ε is normally distributed, then

βˆ ∼ N β, (X 0X )−1 X 0ΣX (X 0X )−1

The problem is that Σ is unknown in general, so this distribution won’t be useful for testing hypotheses.

7.2. THE GLS ESTIMATOR

113

• Without normality, and unconditional on X we still have √ ˆ √ n β−β = n(X 0X )−1 X 0 ε 0 −1 XX = n−1/2 X 0 ε n Define the limiting variance of n−1/2 X 0 ε (supposing a CLT applies) as lim E

n→∞

so we obtain

X 0 εε0 X n

=Ω

√ ˆ d −1 n β − β → N 0, Q−1 ΩQ X X

Summary: OLS with heteroscedasticity and/or autocorrelation is: • unbiased in the same circumstances in which the estimator is unbiased with iid errors • has a different variance than before, so the previous test statistics aren’t valid • is consistent • is asymptotically normally distributed, but with a different limiting covariance matrix. Previous test statistics aren’t valid in this case for this reason. • is inefficient, as is shown below.

7.2. The GLS estimator Suppose Σ were known. Then one could form the Cholesky decomposition P0 P = Σ−1 Here, P is an upper triangular matrix. We have P0 PΣ = In


114

so P0 PΣP0 = P0 , which implies that PΣP0 = In Consider the model P0 y = P0 X β + P0 ε, or, making the obvious definitions, y∗ = X ∗ β + ε ∗ . This variance of ε∗ = Pε is E (Pεε0 P0 ) = PΣP0

= In Therefore, the model y∗ = X ∗ β + ε ∗ E (ε∗ ) = 0

V (ε∗ ) = In satisfies the classical assumptions. The GLS estimator is simply OLS applied to the transformed model: βˆ GLS = (X ∗0 X ∗ )−1 X ∗0 y∗ = (X 0PP0 X )−1 X 0 PP0 y = (X 0Σ−1 X )−1 X 0 Σ−1 y


115

The GLS estimator is unbiased in the same circumstances under which the OLS estimator is unbiased. For example, assuming X is nonstochastic

E (βˆ GLS ) = E (X 0Σ−1 X )−1 X 0 Σ−1 y

= E (X 0Σ−1 X )−1 X 0 Σ−1 (X β + ε = β.

The variance of the estimator, conditional on X can be calculated using βˆ GLS = (X ∗0 X ∗ )−1 X ∗0 y∗ = (X ∗0 X ∗ )−1 X ∗0 (X ∗ β + ε∗ ) = β + (X ∗0 X ∗ )−1 X ∗0 ε∗ so E

βˆ GLS − β

βˆ GLS − β

0

= E (X ∗0 X ∗ )−1 X ∗0 ε∗ ε∗0 X ∗ (X ∗0 X ∗ )−1 = (X ∗0X ∗ )−1 X ∗0 X ∗ (X ∗0 X ∗ )−1

= (X ∗0X ∗ )−1 = (X 0Σ−1 X )−1 Either of these last formulas can be used. • All the previous results regarding the desirable properties of the least squares estimator hold, when dealing with the transformed model, since the transformed model satisfies the classical assumptions.. • Tests are valid, using the previous formulas, as long as we substitute X ∗ in

place of X . Furthermore, any test that involves σ2 can set it to 1. This is

preferable to re-deriving the appropriate formulas.

7.3. FEASIBLE GLS

116

• The GLS estimator is more efficient than the OLS estimator. This is a consequence of the Gauss-Markov theorem, since the GLS estimator is based on a model that satisfies the classical assumptions but the OLS estimator is not. To see this directly, not that (the following needs to be completed) ˆ −Var(βˆ GLS ) = (X 0X )−1 X 0 ΣX (X 0X )−1 − (X 0Σ−1 X )−1 Var(β) = AΣA

0

h i where A = (X 0 X )−1 X 0 − (X 0 Σ−1 X )−1 X 0 Σ−1 . This may not seem obvious, 0

but it is true, as you can verify for yourself. Then noting that AΣA is a 0

quadratic form in a positive definite matrix, we conclude that AΣA is positive semi-definite, and that GLS is efficient relative to OLS. • As one can verify by calculating fonc, the GLS estimator is the solution to the minimization problem βˆ GLS = arg min(y − X β)0Σ−1 (y − X β) so the metric Σ−1 is used to weight the residuals.

7.3. Feasible GLS The problem is that Σ isn’t known usually, so this estimator isn’t available. • Consider the dimension of Σ : it’s an n × n matrix with n2 − n /2 + n = n2 + n /2 unique elements.

• The number of parameters to estimate is larger than n and increases faster

than n. There’s no way to devise an estimator that satisfies a LLN without adding restrictions.

7.3. FEASIBLE GLS

117

• The feasible GLS estimator is based upon making sufficient assumptions regarding the form of Σ so that a consistent estimator can be devised.

Suppose that we parameterize Σ as a function of X and θ, where θ may include β as well as other parameters, so that Σ = Σ(X , θ) where θ is of fixed dimension. If we can consistently estimate θ, we can consistently estimate Σ, as long as Σ(X , θ) is a continuous function of θ (by the Slutsky theorem). In this case, p b = Σ(X , θ) ˆ → Σ Σ(X , θ)

b we obtain the FGLS If we replace Σ in the formulas for the GLS estimator with Σ, estimator. The FGLS estimator shares the same asymptotic properties as GLS. These are (1) Consistency (2) Asymptotic normality (3) Asymptotic efficiency if the errors are normally distributed. (Cramer-Rao). (4) Test procedures are asymptotically valid. In practice, the usual way to proceed is (1) Define a consistent estimator of θ. This is a case-by-case proposition, depending on the parameterization Σ(θ). We’ll see examples below. b = Σ(X , θ) ˆ (2) Form Σ

(3) Calculate the Cholesky factorization Pb = Chol(Σˆ −1 ). (4) Transform the model using

Pˆ 0 y = Pˆ 0 X β + Pˆ 0ε

7.4. HETEROSCEDASTICITY

118

(5) Estimate using OLS on the transformed model. 7.4. Heteroscedasticity Heteroscedasticity is the case where E (εε0 ) = Σ

is a diagonal matrix, so that the errors are uncorrelated, but have different variances. Heteroscedasticity is usually thought of as associated with cross sectional data, though there is absolutely no reason why time series data cannot also be heteroscedastic. Actually, the popular ARCH (autoregressive conditionally heteroscedastic) models explicitly assume that a time series is heteroscedastic. Consider a supply function qi = β1 + β p Pi + βs Si + εi where Pi is price and Si is some measure of size of the ith firm. One might suppose that unobservable factors (e.g., talent of managers, degree of coordination between production units, etc.) account for the error term εi . If there is more variability in these factors for large firms than for small firms, then εi may have a higher variance when Si is high than when it is low. Another example, individual demand. qi = β1 + β p Pi + βm Mi + εi where P is price and M is income. In this case, εi can reflect variations in preferences. There are more possibilities for expression of preferences when one is rich, so it is possible that the variance of εi could be higher when M is high. Add example of group means.


119

7.4.1. OLS with heteroscedastic consistent varcov estimation. Eicker (1967) and White (1980) showed how to modify test statistics to account for heteroscedasticity of unknown form. The OLS estimator has asymptotic distribution √ ˆ d −1 n β − β → N 0, Q−1 ΩQ X X as we’ve already seen. Recall that we defined

lim E

n→∞

X 0 εε0 X n

=Ω

This matrix has dimension K × K and can be consistently estimated, even if we can’t estimate Σ consistently. The consistent estimator, under heteroscedasticity but no au-

tocorrelation is

n b = 1 ∑ x0 xt εˆ 2 Ω n t=1 t t

One can then modify the previous test statistics to obtain tests that are valid when there is heteroscedasticity of unknown form. For example, the Wald test for H0 : Rβ − r = 0 would be

n Rβˆ − r

0

X 0X R n

−1

0 ˆ XX Ω n

−1

R0

!−1

a Rβˆ − r ∼ χ2 (q)

7.4.2. Detection. There exist many tests for the presence of heteroscedasticity. We’ll discuss three methods. Goldfeld-Quandt. The sample is divided in to three parts, with n1 , n2 and n3 observations, where n1 + n2 + n3 = n. The model is estimated using the first and third parts of the sample, separately, so that βˆ 1 and βˆ 3 will be independent. Then we have 0

εˆ 10 εˆ 1 ε1 M 1 ε1 d 2 → χ (n1 − K) = σ2 σ2 and


120

0

εˆ 30 εˆ 3 ε3 M 3 ε3 d 2 → χ (n3 − K) = σ2 σ2 so

εˆ 10 εˆ 1 /(n1 − K) d → F(n1 − K, n3 − K). εˆ 30 εˆ 3 /(n3 − K)

The distributional result is exact if the errors are normally distributed. This test is a two-tailed test. Alternatively, and probably more conventionally, if one has prior ideas about the possible magnitudes of the variances of the observations, one could order the observations accordingly, from largest to smallest. In this case, one would use a conventional one-tailed F-test. Draw picture. • Ordering the observations is an important step if the test is to have any power. • The motive for dropping the middle observations is to increase the difference between the average variance in the subsamples, supposing that there exists heteroscedasticity. This can increase the power of the test. On the other hand, dropping too many observations will substantially increase the variance of the statistics εˆ 10 εˆ 1 and εˆ 30 εˆ 3 . A rule of thumb, based on Monte Carlo experiments is to drop around 25% of the observations. • If one doesn’t have any ideas about the form of the het. the test will probably have low power since a sensible data ordering isn’t available. White’s test. When one has little idea if there exists heteroscedasticity, and no idea of its potential form, the White test is a possibility. The idea is that if there is homoscedasticity, then E (εt2 |xt ) = σ2 , ∀t

so that xt or functions of xt shouldn’t help to explain E (εt2 ). The test works as follows: (1) Since εt isn’t available, use the consistent estimator εˆ t instead.


121

(2) Regress εˆ t2 = σ2 + zt0 γ + vt where zt is a P-vector. zt may include some or all of the variables in xt , as well as other variables. White’s original suggestion was to use xt , plus the set of all unique squares and cross products of variables in xt . (3) Test the hypothesis that γ = 0. The qF statistic in this case is qF =

P (ESSR − ESSU ) /P ESSU / (n − P − 1)

Note that ESSR = T SSU , so dividing both numerator and denominator by this we get qF = (n − P − 1)

R2 1 − R2

Note that this is the R2 or the artificial regression used to test for heteroscedasticity, not the R2 of the original model. An asymptotically equivalent statistic, under the null of no heteroscedasticity (so that R2 should tend to zero), is a

nR2 ∼ χ2 (P). This doesn’t require normality of the errors, though it does assume that the fourth moment of εt is constant, under the null. Question: why is this necessary? • The White test has the disadvantage that it may not be very powerful unless the zt vector is chosen well, and this is hard to do without knowledge of the form of heteroscedasticity. • It also has the problem that specification errors other than heteroscedasticity may lead to rejection.


122

• Note: the null hypothesis of this test may be interpreted as θ = 0 for the

variance model V (εt2 ) = h(α + zt0 θ), where h(·) is an arbitrary function of unknown form. The test is more general than is may appear from the regression that is used.

Plotting the residuals. A very simple method is to simply plot the residuals (or their squares). Draw pictures here. Like the Goldfeld-Quandt test, this will be more informative if the observations are ordered according to the suspected form of the heteroscedasticity.

7.4.3. Correction. Correcting for heteroscedasticity requires that a parametric form for Σ(θ) be supplied, and that a means for estimating θ consistently be determined. The estimation method will be specific to the for supplied for Σ(θ). We’ll consider two examples. Before this, let’s consider the general nature of GLS when there is heteroscedasticity. Multiplicative heteroscedasticity Suppose the model is yt = xt0 β + εt σt2 = E (εt2 ) = zt0 γ but the other classical assumptions hold. In this case εt2 = zt0 γ

δ

δ

+ vt

and vt has mean zero. Nonlinear least squares could be used to estimate γ and δ consistently, were εt observable. The solution is to substitute the squared OLS residuals ˆ εˆ t2 in place of εt2 , since it is consistent by the Slutsky theorem. Once we have γˆ and δ,


123

we can estimate σt2 consistently using σˆ t2

=

δˆ zt0 γˆ

p

→ σt2 .

In the second step, we transform the model by dividing by the standard deviation: yt x0 β εt = t + σˆ t σˆ t σˆ t or yt∗ = xt∗0 β + εt∗ . Asymptotically, this model satisfies the classical assumptions.

• This model is a bit complex in that NLS is required to estimate the model of the variance. A simpler version would be yt

=

xt0 β + εt

σt2 = E (εt2 ) = σ2 ztδ where zt is a single variable. There are still two parameters to be estimated, and the model of the variance is still nonlinear in the parameters. However, the search method can be used in this case to reduce the estimation problem to repeated applications of OLS. • First, we define an interval of reasonable values for δ, e.g., δ ∈ [0, 3]. • Partition this interval into M equally spaced values, e.g., {0, .1, .2, ..., 2.9, 3}.

• For each of these values, calculate the variable ztδm . • The regression

εˆ t2 = σ2 ztδm + vt

is linear in the parameters, conditional on δm , so one can estimate σ2 by OLS.


124

• Save the pairs (σ2m , δm ), and the corresponding ESSm . Choose the pair with the minimum ESSm as the estimate. • Next, divide the model by the estimated standard deviations. • Can refine. Draw picture. • Works well when the parameter to be searched over is low dimensional, as in this case. Groupwise heteroscedasticity A common case is where we have repeated observations on each of a number of economic agents: e.g., 10 years of macroeconomic data on each of a set of countries or regions, or daily observations of transactions of 200 banks. This sort of data is a pooled cross-section time-series model. It may be reasonable to presume that the variance is constant over time within the cross-sectional units, but that it differs across them (e.g., firms or countries of different sizes...). The model is yit = x0it β + εit E (ε2it ) = σ2i , ∀t

where i = 1, 2, ..., G are the agents, and t = 1, 2, ..., n are the observations on each agent. • The other classical assumptions are presumed to hold.

• In this case, the variance σ2i is specific to each agent, but constant over the n observations for that agent. • In this model, we assume that E (εit εis ) = 0. This is a strong assumption that we’ll relax later.

To correct for heteroscedasticity, just estimate each σ2i using the natural estimator: σˆ 2i =

1 n 2 ∑ εˆ it n t=1


125

• Note that we use 1/n here since it’s possible that there are more than n regressors, so n − K could be negative. Asymptotically the difference is unimportant. • With each of these, transform the model as usual: yit x0 β εit = it + σˆ i σˆ i σˆ i Do this for each cross-sectional group. This transformed model satisfies the classical assumptions, asymptotically. 7.4.4. Example: the Nerlove model (again!) Let’s check the Nerlove data for evidence of heteroscedasticity. In what follows, we’re going to use the model with the constant and output coefficient varying across 5 groups, but with the input price coefficients fixed (see Equation 6.8.3 for the rationale behind this). Figure 7.4.1, which is generated by the Octave program GLS/NerloveResiduals.m plots the residuals. We can see pretty clearly that the error variance is larger for small firms than for larger firms. Now let’s try out some tests to formally check for heteroscedasticity. The Octave program GLS/HetTests.m performs the White and Goldfeld-Quandt tests, using the above model. The results are

White’s test

GQ test

Value

p-value

61.903

0.000

Value

p-value

10.886

0.000

All in all, it is very clear that the data are heteroscedastic. That means that OLS estimation is not efficient, and tests of restrictions that ignore heteroscedasticity are not


126

F IGURE 7.4.1. Residuals, Nerlove model, sorted by firm size Regression residuals 1.5

Residuals

1

0.5

0

-0.5

-1

-1.5

0

20

40

60

80

100

120

140

160

valid. The previous tests (CRTS, HOD1 and the Chow test) were calculated assuming homoscedasticity. The Octave program GLS/NerloveRestrictions-Het.m uses the Wald test to check for CRTS and HOD1, but using a heteroscedastic-consistent covariance estimator.1 The results are Testing HOD1

Wald test

Value

p-value

6.161

0.013

Value

p-value

Testing CRTS

1By the way, notice that GLS/NerloveResiduals.m and GLS/HetTests.m use the restricted LS estimator

directly to restrict the fully general model with all coefficients varying to the model with only the constant and the output coefficient varying. But GLS/NerloveRestrictions-Het.m estimates the model by substituting the restrictions into the model. The methods are equivalent, but the second is more convenient and easier to understand.


Wald test

20.169

127

0.001

We see that the previous conclusions are altered - both CRTS is and HOD1 are rejected at the 5% level. Maybe the rejection of HOD1 is due to to Wald test’s tendency to overreject? From the previous plot, it seems that the variance of ε is a decreasing function of output. Suppose that the 5 size groups have different error variances (heteroscedasticity by groups): Var(εi ) = σ2j , where j = 1 if i = 1, 2, ..., 29, etc., as before. The Octave program GLS/NerloveGLS.m estimates the model using GLS (through a transformation of the model so that OLS can be applied). The estimation results are ********************************************************* OLS estimation results Observations 145 R-squared 0.958822 Sigma-squared 0.090800

Results (Het. consistent var-cov estimator)

estimate

st.err.

t-stat.

p-value

constant1

-1.046

1.276

-0.820

0.414

constant2

-1.977

1.364

-1.450

0.149

constant3

-3.616

1.656

-2.184

0.031

constant4

-4.052

1.462

-2.771

0.006


constant5

128

-5.308

1.586

-3.346

0.001

output1

0.391

0.090

4.363

0.000

output2

0.649

0.090

7.184

0.000

output3

0.897

0.134

6.688

0.000

output4

0.962

0.112

8.612

0.000

output5

1.101

0.090

12.237

0.000

labor

0.007

0.208

0.032

0.975

fuel

0.498

0.081

6.149

0.000

-0.460

0.253

-1.818

0.071

capital

*********************************************************

********************************************************* OLS estimation results Observations 145 R-squared 0.987429 Sigma-squared 1.092393

Results (Het. consistent var-cov estimator)

estimate

st.err.

t-stat.

p-value

constant1

-1.580

0.917

-1.723

0.087

constant2

-2.497

0.988

-2.528

0.013

constant3

-4.108

1.327

-3.097

0.002

constant4

-4.494

1.180

-3.808

0.000

constant5

-5.765

1.274

-4.525

0.000


129

output1

0.392

0.090

4.346

0.000

output2

0.648

0.094

6.917

0.000

output3

0.892

0.138

6.474

0.000

output4

0.951

0.109

8.755

0.000

output5

1.093

0.086

12.684

0.000

labor

0.103

0.141

0.733

0.465

fuel

0.492

0.044

11.294

0.000

-0.366

0.165

-2.217

0.028

capital

*********************************************************

Testing HOD1

Wald test

Value

p-value

9.312

0.002

The first panel of output are the OLS estimation results, which are used to consistently estimate the σ2j . The second panel of results are the GLS estimation results. Some comments: • The R2 measures are not comparable - the dependent variables are not the same. The measure for the GLS results uses the transformed dependent variable. One could calculate a comparable R2 measure, but I have not done so. • The differences in estimated standard errors (smaller in general for GLS) can be interpreted as evidence of improved efficiency of GLS, since the OLS standard errors are calculated using the Huber-White estimator. They would not be comparable if the ordinary (inconsistent) estimator had been used.

7.5. AUTOCORRELATION

130

• Note that the previously noted pattern in the output coefficients persists. The nonconstant CRTS result is robust. • The coefficient on capital is now negative and significant at the 3% level. That seems to indicate some kind of problem with the model or the data, or economic theory. • Note that HOD1 is now rejected. Problem of Wald test over-rejecting? Specification error in model? 7.5. Autocorrelation Autocorrelation, which is the serial correlation of the error term, is a problem that is usually associated with time series data, but also can affect cross-sectional data. For example, a shock to oil prices will simultaneously affect all countries, so one could expect contemporaneous correlation of macroeconomic variables across countries. 7.5.1. Causes. Autocorrelation is the existence of correlation across the error term: E (εt εs ) 6= 0,t 6= s.

Why might this occur? Plausible explanations include (1) Lags in adjustment to shocks. In a model such as yt = xt0 β + εt , one could interpret xt0 β as the equilibrium value. Suppose xt is constant over a number of observations. One can interpret εt as a shock that moves the system away from equilibrium. If the time needed to return to equilibrium is long with respect to the observation frequency, one could expect εt+1 to be positive, conditional on εt positive, which induces a correlation.


131

F IGURE 7.5.1. Autocorrelation induced by misspecification

(2) Unobserved factors that are correlated over time. The error term is often assumed to correspond to unobservable factors. If these factors are correlated, there will be autocorrelation. (3) Misspecification of the model. Suppose that the DGP is yt = β0 + β1 xt + β2 xt2 + εt but we estimate yt = β0 + β1 xt + εt The effects are illustrated in Figure 7.5.1.


132

7.5.2. Effects on the OLS estimator. The variance of the OLS estimator is the same as in the case of heteroscedasticity - the standard formula does not apply. The correct formula is given in equation 7.1.1. Next we discuss two GLS corrections for OLS. These will potentially induce inconsistency when the regressors are nonstochastic (see Chapter8) and should either not be used in that case (which is usually the relevant case) or used with caution. The more recommended procedure is discussed in section 7.5.5. 7.5.3. AR(1). There are many types of autocorrelation. We’ll consider two examples. The first is the most commonly encountered case: autoregressive order 1 (AR(1) errors. The model is yt = xt0 β + εt εt = ρεt−1 + ut ut ∼ iid(0, σ2u ) E (εt us ) = 0,t < s

We assume that the model satisfies the other classical assumptions. • We need a stationarity assumption: |ρ| < 1. Otherwise the variance of εt explodes as t increases, so standard asymptotics will not apply. • By recursive substitution we obtain εt = ρεt−1 + ut = ρ (ρεt−2 + ut−1 ) + ut = ρ2 εt−2 + ρut−1 + ut = ρ2 (ρεt−3 + ut−2 ) + ρut−1 + ut


133

In the limit the lagged ε drops out, since ρm → 0 as m → ∞, so we obtain εt =

∞

∑ ρmut−m

m=0

With this, the variance of εt is found as E (εt2 ) = σ2u

∞

∑ ρ2m

m=0

σ2u = 1 − ρ2 • If we had directly assumed that εt were covariance stationary, we could obtain this using 2 V (εt ) = ρ2 E (εt−1 ) + 2ρE (εt−1 ut ) + E (ut2 )

= ρ2V (εt ) + σ2u , so V (εt ) =

σ2u 1 − ρ2

• The variance is the 0th order autocovariance: γ0 = V (εt ) • Note that the variance does not depend on t Likewise, the first order autocovariance γ1 is Cov(εt , εt−1 ) = γs = E ((ρεt−1 + ut ) εt−1 ) =

ρV (εt )

=

ρσ2u 1 − ρ2

• Using the same method, we find that for s < t Cov(εt , εt−s ) = γs =

ρs σ2u 1 − ρ2


134

• The autocovariances don’t depend on t: the process {εt } is covariance stationary

The correlation (in general, for r.v.’s x and y) is defined as corr(x, y) =

cov(x, y) se(x)se(y)

but in this case, the two standard errors are the same, so the s-order autocorrelation ρ s is ρs = ρ s • All this means that the overall matrix Σ has the form  ρ ρ2 · · · ρn−1  1   ρ 1 ρ · · · ρn−2   . σ2u .. ..  .. Σ= . .  2 1−ρ  | {z } ..  . ρ this is the variance   ρn−1 · · · 1 {z | this is the correlation matrix

           

}

So we have homoscedasticity, but elements off the main diagonal are not zero. All of this depends only on two parameters, ρ and σ2u . If we can estimate these consistently, we can apply FGLS.

It turns out that it’s easy to estimate these consistently. The steps are

(1) Estimate the model yt = xt0 β + εt by OLS. (2) Take the residuals, and estimate the model εˆ t = ρεˆ t−1 + ut∗


135

p

Since εˆ t → εt , this regression is asymptotically equivalent to the regression εt = ρεt−1 + ut which satisfies the classical assumptions. Therefore, ρˆ obtained by applying p

OLS to εˆ t = ρεˆ t−1 + ut∗ is consistent. Also, since ut∗ → ut , the estimator σˆ 2u =

1 n ∗ 2 p 2 ∑ (uˆt ) → σu n t=2

ˆ form Σˆ = Σ(σˆ 2u , ρ) ˆ using the previ(3) With the consistent estimators σˆ 2u and ρ, ous structure of Σ, and estimate by FGLS. Actually, one can omit the factor σˆ 2u /(1 − ρ2 ), since it cancels out in the formula βˆ FGLS = X 0 Σˆ −1 X

−1

(X 0Σˆ −1 y).

• One can iterate the process, by taking the first FGLS estimator of β, re-

estimating ρ and σ2u , etc. If one iterates to convergences it’s equivalent to

MLE (supposing normal errors). • An asymptotically equivalent approach is to simply estimate the transformed model ˆ t−1 = (xt − ρx ˆ t−1 )0 β + ut∗ yt − ρy using n − 1 observations (since y0 and x0 aren’t available). This is the method of Cochrane and Orcutt. Dropping the first observation is asymptotically irrelevant, but it can be very important in small samples. One can recuperate the first observation by putting y∗1 = y1 x∗1 = x1

q 1 − ρˆ 2 q 1 − ρˆ 2


136

This somewhat odd-looking result is related to the Cholesky factorization of Σ−1 . See Davidson and MacKinnon, pg. 348-49 for more discussion. Note that the variance of y∗1 is σ2u , asymptotically, so we see that the transformed model will be homoscedastic (and nonautocorrelated, since the u 0 s are uncorrelated with the y0 s, in different time periods. 7.5.4. MA(1). The linear regression model with moving average order 1 errors is yt = xt0 β + εt εt = ut + φut−1 ut ∼ iid(0, σ2u ) E (εt us ) = 0,t < s

In this case, h

2

V (εt ) = γ0 = E (ut + φut−1 ) =

σ2u + φ2 σ2u

=

σ2u (1 + φ2 )

i

Similarly γ1 = E [(ut + φut−1 ) (ut−1 + φut−2 )] = φσ2u and γ2 = [(ut + φut−1 ) (ut−2 + φut−3 )] = 0


so in this case



     2 Σ = σu     

1 + φ2

φ

0

φ

1 + φ2

0 .. .

φ

φ .. .

0

···

137

···

0 .. .

..

.

φ

φ 1 + φ2

Note that the first order autocorrelation is ρ1 =

φσ2u 2 σu (1+φ2 )

=

           

γ1 γ0 φ (1 + φ2 )

=

• This achieves a maximum at φ = 1 and a minimum at φ = −1, and the maximal and minimal autocorrelations are 1/2 and -1/2. Therefore, series that are more strongly autocorrelated can’t be MA(1) processes. Again the covariance matrix has a simple structure that depends on only two parameters. The problem in this case is that one can’t estimate φ using OLS on εˆ t = ut + φut−1 because the ut are unobservable and they can’t be estimated consistently. However, there is a simple way to estimate the parameters. • Since the model is homoscedastic, we can estimate V (εt ) = σ2ε = σ2u (1 + φ2 ) using the typical estimator: 1 n 2 c 2 \ 2 2 σε = σu (1 + φ ) = ∑ εˆ t n t=1


138

• By the Slutsky theorem, we can interpret this as defining an (unidentified) estimator of both σ2u and φ, e.g., use this as

1 n 2 2 c2 (1 + b σ φ ) = ∑ εˆt u n t=1

However, this isn’t sufficient to define consistent estimators of the parameters, since it’s unidentified. • To solve this problem, estimate the covariance of εt and εt−1 using n d2 = 1 d t , εt−1 ) = φσ Cov(ε ∑ εˆt εˆt−1 u n t=2

This is a consistent estimator, following a LLN (and given that the epsilon hats are consistent for the epsilons). As above, this can be interpreted as defining an unidentified estimator: n c2 = 1 φˆ σ ∑ εˆt εˆt−1 u n t=2

• Now solve these two equations to obtain identified (and therefore consistent) estimators of both φ and σ2u . Define the consistent estimator c2 ) ˆ σ Σˆ = Σ(φ, u

following the form we’ve seen above, and transform the model using the Cholesky decomposition. The transformed model satisfies the classical assumptions asymptotically.

7.5.5. Asymptotically valid inferences with autocorrelation of unknown form. See Hamilton Ch. 10, pp. 261-2 and 280-84.


139

When the form of autocorrelation is unknown, one may decide to use the OLS estimator, without correction. We’ve seen that this estimator has the limiting distribution √ ˆ d −1 n β − β → N 0, Q−1 ΩQ X X where, as before, Ω is Ω = lim E n→∞

X 0 εε0 X n

We need a consistent estimate of Ω. Define mt = xt εt (recall that xt is defined as a K × 1 vector). Note that

X 0ε =

h



  i  · · · xn    

x1 x2

n

=

∑ xt εt



ε1   ε2   ..  .    εn

t=1 n

=

∑ mt

t=1

so that 1 Ω = lim E n→∞ n

"

n

∑ mt

t=1

!

n

∑ mt0

t=1

!#

We assume that mt is covariance stationary (so that the covariance between mt and mt−s does not depend on t). Define the v − th autocovariance of mt as 0 Γv = E (mt mt−v ). 0 ) = Γ0 . (show this with an example). In general, we expect that: Note that E (mt mt+v v


140

• mt will be autocorrelated, since εt is potentially autocorrelated: 0 Γv = E (mt mt−v ) 6= 0

Note that this autocovariance does not depend on t, due to covariance stationarity. • contemporaneously correlated ( E (mit m jt ) 6= 0 ), since the regressors in xt will in general be correlated (more on this later). • and heteroscedastic (E (m2it ) = σ2i , which depends upon i ), again since the regressors will have different variances.

While one could estimate Ω parametrically, we in general have little information upon which to base a parametric specification. Recent research has focused on consistent nonparametric estimators of Ω. Now define 1 Ωn = E n

"

n

∑ mt

t=1

!

n

∑ mt0

t=1

!#

We have (show that the following is true, by expanding sum and shifting rows to left) Ωn = Γ0 +

n−2 n−1 1 Γ1 + Γ01 + Γ2 + Γ02 · · · + Γn−1 + Γ0n−1 n n n

The natural, consistent estimator of Γv is

where

1 n 0 . Γbv = ∑ mˆ t mˆ t−v n t=v+1 mˆ t = xt εˆ t


141

(note: one could put 1/(n − v) instead of 1/n here). So, a natural, but inconsistent, estimator of Ωn would be

0 c0 + n − 2 Γ c0 + · · · + 1 Γd d c0 + n − 1 Γ c1 + Γ c2 + Γ ˆn = Γ Ω + Γ n−1 1 2 n−1 n n n n−1 c0 + ∑ n − v Γbv + Γb0v . = Γ v=1 n

This estimator is inconsistent in general, since the number of parameters to estimate is more than the number of observations, and increases more rapidly than n, so information does not build up as n → ∞.

On the other hand, supposing that Γv tends to zero sufficiently rapidly as v tends to

∞, a modified estimator c0 + ˆn=Γ Ω

p

q(n)

∑

v=1

0 b b Γv + Γv ,

where q(n) → ∞ as n → ∞ will be consistent, provided q(n) grows sufficiently slowly.

• The assumption that autocorrelations die off is reasonable in many cases. For example, the AR(1) model with |ρ| < 1 has autocorrelations that die off. • The term

n−v n

can be dropped because it tends to one for v < q(n), given that

q(n) increases slowly relative to n. • A disadvantage of this estimator is that is may not be positive definite. This could cause one to calculate a negative χ2 statistic, for example!

• Newey and West proposed and estimator (Econometrica, 1987) that solves the problem of possible nonpositive definiteness of the above estimator. Their estimator is q(n)

c0 + ∑ ˆn=Γ Ω

v=1

v 0 b b 1− Γv + Γv . q+1


142

This estimator is p.d. by construction. The condition for consistency is that n−1/4 q(n) → 0. Note that this is a very slow rate of growth for q. This estimator is nonparametric - we’ve placed no parametric restrictions on the form of Ω. It is an example of a kernel estimator. p cX = 1 X 0 X to ˆn→ ˆ n and Q Finally, since Ωn has Ω as its limit, Ω Ω. We can now use Ω n

consistently estimate the limiting distribution of the OLS estimator under heteroscedasticity and autocorrelation of unknown form. With this, asymptotically valid tests are constructed in the usual way.

7.5.6. Testing for autocorrelation. Durbin-Watson test The Durbin-Watson test statistic is n (εˆ t − εˆ t−1 )2 ∑t=2 DW = n ˆ2 εt ∑t=1 n 2 εˆ t2 − 2εˆ t εˆ t−1 + εˆ t−1 ∑t=2 = n ˆ2 εt ∑t=1

• The null hypothesis is that the first order autocorrelation of the errors is zero:

H0 : ρ1 = 0. The alternative is of course HA : ρ1 6= 0. Note that the alternative is not that the errors are AR(1), since many general patterns of autocorrelation will have the first order autocorrelation different than zero. For this reason the test is useful for detecting autocorrelation in general. For the same reason, one shouldn’t just assume that an AR(1) model is appropriate when the DW test

rejects the null. • Under the null, the middle term tends to zero, and the other two tend to one, p

so DW → 2.

• Supposing that we had an AR(1) error process with ρ = 1. In this case the p

middle term tends to −2, so DW → 0


143

• Supposing that we had an AR(1) error process with ρ = −1. In this case the p

middle term tends to 2, so DW → 4 • These are the extremes: DW always lies between 0 and 4. • The distribution of the test statistic depends on the matrix of regressors, X , so tables can’t give exact critical values. The give upper and lower bounds, which correspond to the extremes that are possible. See Figure 7.5.2. There are means of determining exact critical values conditional on X . • Note that DW can be used to test for nonlinearity (add discussion). • The DW test is based upon the assumption that the matrix X is fixed in repeated samples. This is often unreasonable in the context of economic time series, which is precisely the context where the test would have application. It is possible to relate the DW test to other test statistics which are valid without strict exogeneity.

Breusch-Godfrey test This test uses an auxiliary regression, as does the White test for heteroscedasticity. The regression is εˆ t = xt0 δ + γ1 εˆ t−1 + γ2 εˆ t−2 + · · · + γP εˆ t−P + vt and the test statistic is the nR2 statistic, just as in the White test. There are P restrictions, so the test statistic is asymptotically distributed as a χ2 (P).

• The intuition is that the lagged errors shouldn’t contribute to explaining the current error if there is no autocorrelation. • xt is included as a regressor to account for the fact that the εˆ t are not independent even if the εt are. This is a technicality that we won’t go into here.


144

F IGURE 7.5.2. Durbin-Watson critical values

• This test is valid even if the regressors are stochastic and contain lagged dependent variables, so it is considerably more useful than the DW test for typical time series data. • The alternative is not that the model is an AR(P), following the argument above. The alternative is simply that some or all of the first P autocorrelations are different from zero. This is compatible with many specific forms of autocorrelation. 7.5.7. Lagged dependent variables and autocorrelation. We’ve seen that the OLS estimator is consistent under autocorrelation, as long as plim Xnε = 0. This will 0

be the case when E (X 0 ε) = 0, following a LLN. An important exception is the case where X contains lagged y0 s and the errors are autocorrelated. A simple example is the


145

case of a single lag of the dependent variable with AR(1) errors. The model is yt = xt0 β + yt−1 γ + εt εt = ρεt−1 + ut Now we can write

0 E (yt−1 εt ) = E (xt−1 β + yt−2 γ + εt−1 )(ρεt−1 + ut )

6= 0

2 ) which is clearly nonzero. In this case E (X 0 ε) 6= 0, since one of the terms is E (ρεt−1

and therefore plim Xnε 6= 0. Since 0

plimβˆ = β + plim

X 0ε n

the OLS estimator is inconsistent in this case. One needs to estimate by instrumental variables (IV), which we’ll get to later.

7.5.8. Examples. Nerlove model, yet again. The Nerlove model uses cross-sectional data, so one may not think of performing tests for autocorrelation. However, specification error can induce autocorrelated errors. Consider the simple Nerlove model lnC = β1 + β2 ln Q + β3 ln PL + β4 ln PF + β5 ln PK + ε and the extended Nerlove model j

j

lnC = β1 + β2 ln Q + β3 ln PL + β4 ln PF + β5 ln PK + ε.


146

F IGURE 7.6.1. Residuals of simple Nerlove model 2

Residuals Quadratic fit to Residuals

1.5

1

0.5

0

-0.5

-1

0

1

2

3

4

5

6

7

8

9

10

We have seen evidence that the extended model is preferred. So if it is in fact the proper model, the simple model is misspecified. Let’s check if this misspecification might induce autocorrelated errors. The Octave program GLS/NerloveAR.m estimates the simple Nerlove model, and plots the residuals as a function of ln Q, and it calculates a Breusch-Godfrey test statistic. The residual plot is in Figure 7.6.1 , and the test results are:

Breusch-Godfrey test

Value

p-value

34.930

0.000

Clearly, there is a problem of autocorrelated residuals. E XERCISE 7.6. Repeat the autocorrelation tests using the extended Nerlove model (Equation ??) to see the problem is solved.


147

Klein model. Klein’s Model I is a simple macroeconometric model. One of the equations in the model explains consumption (C) as a function of profits (P), both current and lagged, as well as the sum of wages in the private sector (W p ) and wages in the government sector (W g ). Have a look at the README file for this data set. This gives the variable names and other information. Consider the model p

g

Ct = α0 + α1 Pt + α2 Pt−1 + α3 (Wt +Wt ) + ε1t The Octave program GLS/Klein.m estimates this model by OLS, plots the residuals, and performs the Breusch-Godfrey test, using 1 lag of the residuals. The estimation and test results are:


Results (Ordinary var-cov estimator)

estimate

st.err.

t-stat.

p-value

Constant

16.237

1.303

12.464

0.000

Profits

0.193

0.091

2.115

0.049

Lagged Profits

0.090

0.091

0.992

0.335

Wages

0.796

0.040

19.933

0.000


148

F IGURE 7.6.2. OLS residuals, Klein consumption equation Regression residuals 2

Residuals

1.5 1 0.5 0 -0.5 -1 -1.5 -2 -2.5

0

5

10

15

20

25

*********************************************************


Value

p-value

1.539

0.215

and the residual plot is in Figure 7.6.2. The test does not reject the null of nonautocorrelatetd errors, but we should remember that we have only 21 observations, so power is likely to be fairly low. The residual plot leads me to suspect that there may be autocorrelation - there are some significant runs below and above the x-axis. Your opinion may differ. Since it seems that there may be autocorrelation, lets’s try an AR(1) correction. The Octave program GLS/KleinAR1.m estimates the Klein consumption equation assuming that the errors follow the AR(1) pattern. The results, with the Breusch-Godfrey test for remaining autocorrelation are:


149


Results (Ordinary var-cov estimator)

estimate

st.err.

t-stat.

p-value

Constant

16.992

1.492

11.388

0.000

Profits

0.215

0.096

2.232

0.039

Lagged Profits

0.076

0.094

0.806

0.431

Wages

0.774

0.048

16.234

0.000

*********************************************************


Value

p-value

2.129

0.345

• The test is farther away from the rejection region than before, and the residual plot is a bit more favorable for the hypothesis of nonautocorrelated residuals, IMHO. For this reason, it seems that the AR(1) correction might have improved the estimation. • Nevertheless, there has not been much of an effect on the estimated coefficients nor on their estimated standard errors. This is probably because the estimated AR(1) coefficient is not very large (around 0.2)

EXERCISES

150

• The existence or not of autocorrelation in this model will be important later, in the section on simultaneous equations. Exercises

EXERCISES

151

(1) Comparing the variances of the OLS and GLS estimators, I claimed that the following holds: (2) ˆ −Var(βˆ GLS ) = AΣA0 Var(β) Verify that this is true. (3) Show that the GLS estimator can be defined as βˆ GLS = arg min(y − X β)0Σ−1 (y − X β) (4) The limiting distribution of the OLS estimator with heteroscedasticity of unknown form is

√ ˆ d −1 −1 n β − β → N 0, QX ΩQX ,

where

lim E

n→∞

Explain why

X 0 εε0 X n

=Ω

n b = 1 ∑ x0 xt εˆ 2 Ω n t=1 t t

is a consistent estimator of this matrix.

(5) Define the v − th autocovariance of a covariance stationary process mt , where E(mt = 0) as 0 Γv = E (mt mt−v ). 0 ) = Γ0 . Show that E (mt mt+v v

(6) For the Nerlove model j

j

lnC = β1 + β2 ln Q + β3 ln PL + β4 ln PF + β5 ln PK + ε

EXERCISES

152

assume that V (εt |xt ) = α + αQ ln Q. Exercises (a) Calculate the FGLS estimator and interpret the estimation results. (b) Test the transformed model to check whether it appears to satisfy homoscedasticity.

CHAPTER 8

Stochastic regressors Up to now we have treated the regressors as fixed, which is clearly unrealistic. Now we will assume they are random. There are several ways to think of the problem. First, if we are interested in an analysis conditional on the explanatory variables, then it is irrelevant if they are stochastic or not, since conditional on the values of they regressors take on, they are nonstochastic, which is the case already considered. • In cross-sectional analysis it is usually reasonable to make the analysis conditional on the regressors. • In dynamic models, where yt may depend on yt−1 , a conditional analysis is not sufficiently general, since we may want to predict into the future many periods out, so we need to consider the behavior of βˆ and the relevant test statistics unconditional on X . The model we’ll deal will involve a combination of the following assumptions Linearity: the model is a linear function of the parameter vector β0 : yt = xt0 β0 + εt , or in matrix form,

where y is n × 1, X = formable.

x1 x2

y = X β0 + ε, 0 · · · xn , where xt is K × 1, and β0 and ε are con-

Stochastic, linearly independent regressors 153

8.1. CASE 1

154

X has rank K with probability 1 X is stochastic 1 0 nX X

limn→∞ Pr

= QX = 1, where QX is a finite positive definite matrix.

Central limit theorem d

n−1/2 X 0ε → N(0, QX σ20 )

Normality (Optional): ε|X ∼ N(0, σ2 In ): ε is normally distributed Strongly exogenous regressors: (8.0.1)

E (εt |X) = 0, ∀t

Weakly exogenous regressors: (8.0.2)

E(εt |xt ) = 0, ∀t

In both cases, xt0 β is the conditional mean of yt given xt : E(yt |xt ) = xt0 β

8.1. Case 1 Normality of ε, strongly exogenous regressors In this case, βˆ = β0 + (X 0 X )−1 X 0ε ˆ ) = β0 + (X 0 X )−1 X 0 E (ε|X ) E (β|X = β0 ˆ = β, unconditional on X . Likewise, and since this holds for all X , E(β) ˆ ∼ N β, (X 0X )−1 σ2 β|X 0

8.2. CASE 2

155

• If the density of X is dµ(X ), the marginal density of βˆ is obtained by multiplying the conditional density by dµ(X ) and integrating over X . Doing this ˆ in small samples. leads to a nonnormal density for β, • However, conditional on X , the usual test statistics have the t, F and χ 2 distributions. Importantly, these distributions don’t depend on X , so when marginalizing to obtain the unconditional distribution, nothing changes. The tests are valid in small samples. • Summary: When X is stochastic but strongly exogenous and ε is normally distributed: (1) βˆ is unbiased (2) βˆ is nonnormally distributed (3) The usual test statistics have the same distribution as with nonstochastic X. (4) The Gauss-Markov theorem still holds, since it holds conditionally on X , and this is true for all X . (5) Asymptotic properties are treated in the next section.

8.2. Case 2 ε nonnormally distributed, strongly exogenous regressors The unbiasedness of βˆ carries through as before. However, the argument regarding test statistics doesn’t hold, due to nonnormality of ε. Still, we have βˆ = β0 + (X 0 X )−1 X 0 ε 0 −1 0 XX Xε = β0 + n n

8.2. CASE 2

Now

by assumption, and

X 0X n

−1

156

p

→ Q−1 X

X 0 ε n−1/2 X 0 ε p = √ →0 n n

since the numerator converges to a N(0, QX σ2 ) r.v. and the denominator still goes to infinity. We have unbiasedness and the variance disappearing, so, the estimator is consistent: p βˆ → β0 .

Considering the asymptotic distribution √ X 0 X −1 X 0 ε √ ˆ n β − β0 = n n n 0 −1 XX n−1/2 X 0 ε = n so

√ ˆ d 2 n β − β0 → N(0, Q−1 X σ0 )

directly following the assumptions. Asymptotic normality of the estimator still holds. Since the asymptotic results on all test statistics only require this, all the previous asymptotic results on test statistics are also valid in this case. • Summary: Under strongly exogenous regressors, with ε normal or nonnormal, βˆ has the properties: (1) Unbiasedness (2) Consistency (3) Gauss-Markov theorem holds, since it holds in the previous case and doesn’t depend on normality. (4) Asymptotic normality

8.4. WHEN ARE THE ASSUMPTIONS REASONABLE?

157

(5) Tests are asymptotically valid, but are not valid in small samples. 8.3. Case 3 Weakly exogenous regressors An important class of models are dynamic models, where lagged dependent variables have an impact on the current value. A simple version of these models that captures the important points is yt =

zt0 α +

p

∑ γsyt−s + εt

s=1

= xt0 β + εt where now xt contains lagged dependent variables. Clearly, even with E(εt |xt ) = 0, X

and ε are not uncorrelated, so one can’t show unbiasedness. For example, E (εt−1 xt ) 6= 0

since xt contains yt−1 (which is a function of εt−1 ) as an element. • This fact implies that all of the small sample properties such as unbiasedness, Gauss-Markov theorem, and small sample validity of test statistics do not hold in this case. Recall Figure 3.7.2. This is a case of weakly exogenous regressors, and we see that the OLS estimator is biased in this case. • Nevertheless, under the above assumptions, all asymptotic properties continue to hold, using the same arguments as before. 8.4. When are the assumptions reasonable? The two assumptions we’ve added are (1) limn→∞ Pr 1n X 0 X = QX = 1, a QX finite positive definite matrix. d

(2) n−1/2 X 0 ε → N(0, QX σ20 )


158

The most complicated case is that of dynamic models, since the other cases can be treated as nested in this case. There exist a number of central limit theorems for dependent processes, many of which are fairly technical. We won’t enter into details (see Hamilton, Chapter 7 if you’re interested). A main requirement for use of standard asymptotics for a dependent sequence 1 n {st } = { ∑ zt } n t=1 to converge in probability to a finite limit is that zt be stationary, in some sense. • Strong stationarity requires that the joint distribution of the set {zt , zt+s , zt−q , ...} not depend on t. • Covariance (weak) stationarity requires that the first and second moments of this set not depend on t. • An example of a sequence that doesn’t satisfy this is an AR(1) process with a unit root (a random walk): xt = xt−1 + εt εt ∼ IIN(0, σ2) One can show that the variance of xt depends upon t in this case. Stationarity prevents the process from trending off to plus or minus infinity, and prevents cyclical behavior which would allow correlations between far removed zt znd zs to be high. Draw a picture here. • In summary, the assumptions are reasonable when the stochastic conditioning variables have variances that are finite, and are not too strongly dependent.


159

The AR(1) model with unit root is an example of a case where the dependence is too strong for standard asymptotics to apply. • The econometrics of nonstationary processes has been an active area of research in the last two decades. The standard asymptotics don’t apply in this case. This isn’t in the scope of this course.

EXERCISES

160

Exercises (1) Show that for two random variables A and B, if E(A|B) = 0, then E (A f (B)) = 0. How is this used in the Gauss-Markov theorem? (2) If it possible for an AR(1) model for time series data, e.g., yt = 0 + 0.9yt−1 + εt satisfy weak exogeneity? Strong exogeneity? Discuss.

CHAPTER 9

Data problems In this section well consider problems associated with the regressor matrix: collinearity, missing observation and measurement error.

9.1. Collinearity Collinearity is the existence of linear relationships amongst the regressors. We can always write λ1 x 1 + λ 2 x 2 + · · · + λ K x K + v = 0 where xi is the ith column of the regressor matrix X , and v is an n × 1 vector. In the case that there exists collinearity, the variation in v is relatively small, so that there is an approximately exact linear relation between the regressors. • “relative” and “approximate” are imprecise, so it’s difficult to define when collinearilty exists. In the extreme, if there are exact linear relationships (every element of v equal) then ρ(X ) < K, so ρ(X 0X ) < K, so X 0 X is not invertible and the OLS estimator is not uniquely defined. For example, if the model is yt = β1 + β2 x2t + β3 x3t + εt x2t = α1 + α2 x3t 161

9.1. COLLINEARITY

162

then we can write yt = β1 + β2 (α1 + α2 x3t ) + β3 x3t + εt = β1 + β2 α1 + β2 α2 x3t + β3 x3t + εt = (β1 + β2 α1 ) + (β2 α2 + β3 ) x3t = γ1 + γ2 x3t + εt • The γ0 s can be consistently estimated, but since the γ0 s define two equations in three β0 s, the β0 s can’t be consistently estimated (there are multiple values of β

that solve the fonc). The β0 s are unidentified in the case of perfect collinearity. • Perfect collinearity is unusual, except in the case of an error in construction of the regressor matrix, such as including the same regressor twice. Another case where perfect collinearity may be encountered is with models with dummy variables, if one is not careful. Consider a model of rental price (y i ) of an apartment. This could depend factors such as size, quality etc., collected in x i , as well as on the location of the apartment. Let Bi = 1 if the ith apartment is in Barcelona, Bi = 0 otherwise. Similarly, define Gi , Ti and Li for Girona, Tarragona and Lleida. One could use a model such as yi = β1 + β2 Bi + β3 Gi + β4 Ti + β5 Li + x0i γ + εi In this model, Bi + Gi + Ti + Li = 1, ∀i, so there is an exact relationship between these variables and the column of ones corresponding to the constant. One must either drop the constant, or one of the qualitative variables.

9.1.1. A brief aside on dummy variables. Introduce a brief discussion of dummy variables here.

9.1. COLLINEARITY

163

F IGURE 9.1.1. s(β) when there is no collinearity

6 4

60 55 50 45 40 35 30 25 20 15

2 0 -2 -4

-6

-4

-2

0

2

4

6

-6

9.1.2. Back to collinearity. The more common case, if one doesn’t make mistakes such as these, is the existence of inexact linear relationships, i.e., correlations between the regressors that are less than one in absolute value, but not zero. The basic problem is that when two (or more) variables move together, it is difficult to determine their separate influences. This is reflected in imprecise estimates, i.e., estimates with high variances. With economic data, collinearity is commonly encountered, and is often a severe problem. When there is collinearity, the minimizing point of the objective function that defines the OLS estimator (s(β), the sum of squared errors) is relatively poorly defined. This is seen in Figures 9.1.1 and 9.1.2. To see the effect of collinearity on variances, partition the regressor matrix as X=

h

x W

i

9.1. COLLINEARITY

164

F IGURE 9.1.2. s(β) when there is collinearity

6 4

100 90 80 70 60 50 40 30 20

2 0 -2 -4

-6

-4

-2

0

2

4

6

-6

where x is the first column of X (note: we can interchange the columns of X isf we like, so there’s no loss of generality in considering the first column). Now, the variance of ˆ under the classical assumptions, is β, ˆ = X 0X V (β) Using the partition,



X 0X = 

−1

x0 x

x0W

W 0x

W 0W

and following a rule for partitioned inversion, X 0X

−1 1,1

σ2  

−1 x0 x − x0W (W 0W )−1W 0 x −1 01 0 0 0 = x In −W (W W ) W x =

=

ESSx|W

−1

9.1. COLLINEARITY

165

where by ESSx|W we mean the error sum of squares obtained from the regression x = W λ + v. Since R2 = 1 − ESS/T SS, we have ESS = T SS(1 − R2 ) so the variance of the coefficient corresponding to x is V (βˆ x ) =

σ2 T SSx (1 − R2x|W )

We see three factors influence the variance of this coefficient. It will be high if

(1) σ2 is large (2) There is little variation in x. Draw a picture here. (3) There is a strong linear relationship between x and the other regressors, so that W can explain the movement in x well. In this case, R2x|W will be close to 1. As R2x|W → 1,V (βˆ x ) → ∞. The last of these cases is collinearity. Intuitively, when there are strong linear relations between the regressors, it is difficult to determine the separate influence of the regressors on the dependent variable. This can be seen by comparing the OLS objective function in the case of no correlation between regressors with the objective function with correlation between the regressors. See the figures nocollin.ps (no correlation) and collin.ps (correlation), available on the web site.

9.1. COLLINEARITY

166

9.1.3. Detection of collinearity. The best way is simply to regress each explanatory variable in turn on the remaining regressors. If any of these auxiliary regressions has a high R2 , there is a problem of collinearity. Furthermore, this procedure identifies which parameters are affected. • Sometimes, we’re only interested in certain parameters. Collinearity isn’t a problem if it doesn’t affect what we’re interested in estimating. An alternative is to examine the matrix of correlations between the regressors. High correlations are sufficient but not necessary for severe collinearity. Also indicative of collinearity is that the model fits well (high R2 ), but none of the variables is significantly different from zero (e.g., their separate influences aren’t well determined). In summary, the artificial regressions are the best approach if one wants to be careful.

9.1.4. Dealing with collinearity. More information Collinearity is a problem of an uninformative sample. The first question is: is all the available information being used? Is more data available? Are there coefficient restrictions that have been neglected? Picture illustrating how a restriction can solve problem of perfect collinearity. Stochastic restrictions and ridge regression Supposing that there is no more data or neglected restrictions, one possibility is to change perspectives, to Bayesian econometrics. One can express prior beliefs regarding the coefficients using stochastic restrictions. A stochastic linear restriction would be something of the form Rβ = r + v

9.1. COLLINEARITY

167

where R and r are as in the case of exact linear restrictions, but v is a random vector. For example, the model could be y = Xβ + ε  

ε v

Rβ = r + v    ∼ N

0 0

  ,

σ2ε In

0n×q

0q×n

σ2v Iq

 

This sort of model isn’t in line with the classical interpretation of parameters as constants: according to this interpretation the left hand side of Rβ = r + v is constant but the right is random. This model does fit the Bayesian perspective: we combine information coming from the model and the data, summarized in y = Xβ + ε ε ∼ N(0, σ2ε In ) with prior beliefs regarding the distribution of the parameter, summarized in Rβ ∼ N(r, σ2v Iq ) Since the sample is random it is reasonable to suppose that E (εv0 ) = 0, which is the last piece of information in the specification. How can you estimate using this model? The solution is to treat the restrictions as artificial data. Write       ε X y β+   = v R r

This model is heteroscedastic, since σ2ε 6= σ2v . Define the prior precision k = σε /σv . This expresses the degree of belief in the restriction relative to the variability of the

9.1. COLLINEARITY

168

data. Supposing that we specify k, then the model  

y kr





=

X kR





β+

ε kv

 

is homoscedastic and can be estimated by OLS. Note that this estimator is biased. It is consistent, however, given that k is a fixed constant, even if the restriction is false (this is in contrast to the case of false exact restrictions). To see this, note that there are Q restrictions, where Q is the number of rows of R. As n → ∞, these Q artificial observations have no weight in the objective function, so the estimator has the same limiting objective function as the OLS estimator, and is therefore consistent. To motivate the use of stochastic restrictions, consider the expectation of the squared ˆ length of β: ˆ0ˆ

E (β β) = E

β+ X X 0

−1

Xε 0

0

β+ X X 0

= β0 β + E ε0 X (X 0X )−1 (X 0 X )−1 X 0 ε = β0 β + Tr X 0 X = β β+σ 0

2

K

−1

∑ λi(the

σ2

−1

Xε 0

trace is the sum of eigenvalues)

i=1

> β0 β + λmax(X 0 X −1 ) σ2 (the eigenvalues are all positive, sinceX 0 X is p.d. so ˆ > β0 β + E (βˆ 0 β)

σ2 λmin(X 0 X)

where λmin(X 0 X) is the minimum eigenvalue of X 0 X (which is the inverse of the maximum eigenvalue of (X 0 X )−1 ). As collinearity becomes worse and worse, X 0 X becomes more nearly singular, so λmin(X 0 X) tends to zero (recall that the determinant is the prodˆ tends to infinite. On the other hand, β0 β is finite. uct of the eigenvalues) and E (βˆ 0 β)

9.1. COLLINEARITY

169

Now considering the restriction IK β = 0 + v. With this restriction the model becomes

 

and the estimator is 

y 0





h

βˆ ridge =  X 0 kIK =

X

=

X 0 X + k2 IK

kIK

i

 

−1





β+ X

kIK

X 0y

ε kv

−1 

h

 

X 0 IK

i

 

y 0

 

This is the ordinary ridge regression estimator. The ridge regression estimator can be seen to add k2 IK , which is nonsingular, to X 0 X , which is more and more nearly singular as collinearity becomes worse and worse. As k → ∞, the restrictions tend to β = 0, that is, the coefficients are shrunken toward zero. Also, the estimator tends to βˆ ridge = X 0 X + k2 IK

−1

X 0 y → k2 IK

−1

X 0y =

X 0y →0 k2

so βˆ 0ridge βˆ ridge → 0. This is clearly a false restriction in the limit, if our original model is at al sensible. There should be some amount of shrinkage that is in fact a true restriction. The problem is to determine the k such that the restriction is correct. The interest in ridge regression centers on the fact that it can be shown that there exists a k such that MSE(βˆ ridge ) < βˆ OLS . The problem is that this k depends on β and σ2 , which are unknown. The ridge trace method plots βˆ 0ridge βˆ ridge as a function of k, and chooses the value of k that “artistically” seems appropriate (e.g., where the effect of increasing k dies off). Draw picture here. This means of choosing k is obviously subjective. This is not

9.2. MEASUREMENT ERROR

170

a problem from the Bayesian perspective: the choice of k reflects prior beliefs about the length of β. In summary, the ridge estimator offers some hope, but it is impossible to guarantee that it will outperform the OLS estimator. Collinearity is a fact of life in econometrics, and there is no clear solution to the problem.

9.2. Measurement error Measurement error is exactly what it says, either the dependent variable or the regressors are measured with error. Thinking about the way economic data are reported, measurement error is probably quite prevalent. For example, estimates of growth of GDP, inflation, etc. are commonly revised several times. Why should the last revision necessarily be correct? 9.2.1. Error of measurement of the dependent variable. Measurement errors in the dependent variable and the regressors have important differences. First consider error in measurement of the dependent variable. The data generating process is presumed to be y∗ = X β + ε y = y∗ + v vt ∼ iid(0, σ2v ) where y∗ is the unobservable true dependent variable, and y is what is observed. We assume that ε and v are independent and that y∗ = X β + ε satisfies the classical assumptions. Given this, we have y + v = Xβ + ε


171

so y = Xβ + ε − v = Xβ + ω ωt ∼ iid(0, σ2ε + σ2v ) • As long as v is uncorrelated with X , this model satisfies the classical assumptions and can be estimated by OLS. This type of measurement error isn’t a problem, then.

9.2.2. Error of measurement of the regressors. The situation isn’t so good in this case. The DGP is yt = xt∗0 β + εt xt = xt∗ + vt vt ∼ iid(0, Σv) where Σv is a K × K matrix. Now X ∗ contains the true, unobserved regressors, and X is what is observed. Again assume that v is independent of ε, and that the model y = X ∗ β + ε satisfies the classical assumptions. Now we have yt = (xt − vt )0 β + εt = xt0 β − vt0 β + εt = xt0 β + ωt


172

The problem is that now there is a correlation between xt and ωt , since E (xt ωt ) = E (xt∗ + vt ) −vt0 β + εt

= −Σv β

where Σv = E vt vt0 .

Because of this correlation, the OLS estimator is biased and inconsistent, just as in the case of autocorrelated errors with lagged dependent variables. In matrix notation, write the estimated model as y = Xβ + ω We have that βˆ =

X 0X n

−1

X 0y n

and

X 0X plim n

−1

= plim

(X ∗0 +V 0 ) (X ∗ +V ) n

= (QX ∗ + Σv )−1

since X ∗ and V are independent, and plim

V 0V n

= lim E

1 n ∑ vt vt0 n t=1

= Σv Likewise,

X 0y plim n

= plim

(X ∗0 +V 0 ) (X ∗ β + ε) n

= QX ∗ β

9.3. MISSING OBSERVATIONS

173

so plimβˆ = (QX ∗ + Σv )−1 QX ∗ β So we see that the least squares estimator is inconsistent when the regressors are measured with error. • A potential solution to this problem is the instrumental variables (IV) estimator, which we’ll discuss shortly.

9.3. Missing observations Missing observations occur quite frequently: time series data may not be gathered in a certain year, or respondents to a survey may not answer all questions. We’ll consider two cases: missing observations on the dependent variable and missing observations on the regressors. 9.3.1. Missing observations on the dependent variable. In this case, we have y = Xβ + ε or

 

y1 y2





=

X1 X2





β+

ε1 ε2

 

where y2 is not observed. Otherwise, we assume the classical assumptions hold. • A clear alternative is to simply estimate using the compete observations y1 = X1 β + ε1 Since these observations satisfy the classical assumptions, one could estimate by OLS.


174

• The question remains whether or not one could somehow replace the unobserved y2 by a predictor, and improve over OLS in some sense. Let yˆ2 be the predictor of y2 . Now

  0  −1  0      X1 y X X  1  1   1   βˆ =    X2 yˆ2 X2 X2  =

−1 0 0 X1 y1 + X20 yˆ2 X1 X1 + X20 X2

Recall that the OLS fonc are

X 0 X βˆ = X 0 y so if we regressed using only the first (complete) observations, we would have X10 X1 βˆ 1 = X10 y1. Likewise, an OLS regression using only the second (filled in) observations would give X20 X2 βˆ 2 = X20 yˆ2 . Substituting these into the equation for the overall combined estimator gives βˆ = =

−1 X10 X1 + X20 X2 X10 X1 + X20 X2

−1

≡ Aβˆ 1 + (IK − A)βˆ 2

h

X10 X1 βˆ 1 + X20 X2 βˆ 2

i

−1 0 X10 X1 βˆ 1 + X10 X1 + X20 X2 X2 X2 βˆ 2

where −1 0 A ≡ X10 X1 + X20 X2 X1 X1


175

and we use 0 −1 0 −1 0 X1 X1 + X20 X2 X2 X2 = X10 X1 + X20 X2 X1 X1 + X20 X2 − X10 X1 −1 0 = IK − X10 X1 + X20 X2 X1 X1 = IK − A.

Now,

ˆ = Aβ + (IK − A)E βˆ 2 E (β)

and this will be unbiased only if E βˆ 2 = β.

• The conclusion is the this filled in observations alone would need to define an unbiased estimator. This will be the case only if yˆ2 = X2 β + εˆ 2 where εˆ 2 has mean zero. Clearly, it is difficult to satisfy this condition without knowledge of β. • Note that putting yˆ2 = y¯1 does not satisfy the condition and therefore leads to a biased estimator.

E XERCISE 13. Formally prove this last statement.

• One possibility that has been suggested (see Greene, page 275) is to estimate β using a first round estimation using only the complete observations βˆ 1 = (X10 X1 )−1 X10 y1


176

then use this estimate, βˆ 1 ,to predict y2 : yˆ2 = X2 βˆ 1 = X2 (X10 X1 )−1 X10 y1 Now, the overall estimate is a weighted average of βˆ 1 and βˆ 2 , just as above, but we have βˆ 2 = (X20 X2 )−1 X20 yˆ2 = (X20 X2 )−1 X20 X2 βˆ 1 = βˆ 1 This shows that this suggestion is completely empty of content: the final estimator is the same as the OLS estimator using only the complete observations.

9.3.2. The sample selection problem. In the above discussion we assumed that the missing observations are random. The sample selection problem is a case where the missing observations are not random. Consider the model yt∗ = xt0 β + εt which is assumed to satisfy the classical assumptions. However, yt∗ is not always observed. What is observed is yt defined as yt = yt∗ if yt∗ ≥ 0 Or, in other words, yt∗ is missing when it is less than zero.


177

F IGURE 9.3.1. Sample selection bias 25

Data True Line Fitted Line

20

15

10

5

0

-5

-10

0

2

4

6

8

10

The difference in this case is that the missing values are not random: they are correlated with the xt . Consider the case y∗ = x + ε with V (ε) = 25, but using only the observations for which y∗ > 0 to estimate. Figure 9.3.1 illustrates the bias. The Octave program is sampsel.m 9.3.3. Missing observations on the regressors. Again the model is  

y1 y2





=

X1 X2





β+

ε1 ε2

 

but we assume now that each row of X2 has an unobserved component(s). Again, one could just estimate using the complete observations, but it may seem frustrating to have to drop observations simply because of a single missing variable. In general,


178

if the unobserved X2 is replaced by some prediction, X2∗ , then we are in the case of errors of observation. As before, this means that the OLS estimator is biased when X2∗ is used instead of X2 . Consistency is salvaged, however, as long as the number of missing observations doesn’t increase with n. • Including observations that have missing values replaced by ad hoc values can be interpreted as introducing false stochastic restrictions. In general, this introduces bias. It is difficult to determine whether MSE increases or decreases. Monte Carlo studies suggest that it is dangerous to simply substitute the mean, for example. • In the case that there is only one regressor other than the constant, subtitution of x¯ for the missing xt does not lead to bias. This is a special case that doesn’t hold for K > 2. E XERCISE 14. Prove this last statement. • In summary, if one is strongly concerned with bias, it is best to drop observations that have missing components. There is potential for reduction of MSE through filling in missing elements with intelligent guesses, but this could also increase MSE.

EXERCISES

179

Exercises (1) Consider the Nerlove model j

j

lnC = β1 + β2 ln Q + β3 ln PL + β4 ln PF + β5 ln PK + ε When this model is estimated by OLS, some coefficients are not significant. This may be due to collinearity. Exercises (a) Calculate the correlation matrix of the regressors. (b) Perform artificial regressions to see if collinearity is a problem. (c) Apply the ridge regression estimator. Exercises (i) Plot the ridge trace diagram (ii) Check what happens as k goes to zero, and as k becomes very large.

CHAPTER 10

Functional form and nonnested tests Though theory often suggests which conditioning variables should be included, and suggests the signs of certain derivatives, it is usually silent regarding the functional form of the relationship between the dependent variable and the regressors. For example, considering a cost function, one could have a Cobb-Douglas model β

β

c = Aw1 1 w2 2 qβq eε This model, after taking logarithms, gives ln c = β0 + β1 ln w1 + β2 ln w2 + βq ln q + ε where β0 = ln A. Theory suggests that A > 0, β1 > 0, β2 > 0, β3 > 0. This model isn’t compatible with a fixed cost of production since c = 0 when q = 0. Homogeneity of degree one in input prices suggests that β1 + β2 = 1, while constant returns to scale implies βq = 1. While this model may be reasonable in some cases, an alternative √ √ √ √ c = β 0 + β 1 w1 + β 2 w2 + β q q + ε may be just as plausible. Note that

√

x and ln(x) look quite alike, for certain values

of the regressors, and up to a linear transformation, so it may be difficult to choose between these models.

180

10.1. FLEXIBLE FUNCTIONAL FORMS

181

The basic point is that many functional forms are compatible with the linear-inparameters model, since this model can incorporate a wide variety of nonlinear transformations of the dependent variable and the regressors. For example, suppose that g(·) is a real valued function and that x(·) is a K− vector-valued function. The following model is linear in the parameters but nonlinear in the variables: xt = x(zt ) yt = xt0 β + εt There may be P fundamental conditioning variables zt , but there may be K regressors, where K may be smaller than, equal to or larger than P. For example, xt could include squares and cross products of the conditioning variables in zt .

10.1. Flexible functional forms Given that the functional form of the relationship between the dependent variable and the regressors is in general unknown, one might wonder if there exist parametric models that can closely approximate a wide variety of functional relationships. A “Diewert-Flexible” functional form is defined as one such that the function, the vector of first derivatives and the matrix of second derivatives can take on an arbitrary value at a single data point. Flexibility in this sense clearly requires that there be at least K = 1 + P + P2 − P /2 + P

free parameters: one for each independent effect that we wish to model. Suppose that the model is y = g(x) + ε


182

A second-order Taylor’s series expansion (with remainder term) of the function g(x) about the point x = 0 is g(x) = g(0) + x0 Dx g(0) +

x0 D2x g(0)x +R 2

Use the approximation, which simply drops the remainder term, as an approximation to g(x) : g(x) ' gK (x) = g(0) + x0 Dx g(0) +

x0 D2x g(0)x 2

As x → 0, the approximation becomes more and more exact, in the sense that g K (x) → g(x), Dx gK (x) → Dx g(x) and D2x gK (x) → D2x g(x). For x = 0, the approximation is

exact, up to the second order. The idea behind many flexible functional forms is to note that g(0), Dx g(0) and D2x g(0) are all constants. If we treat them as parameters, the approximation will have exactly enough free parameters to approximate the function g(x), which is of unknown form, exactly, up to second order, at the point x = 0. The model is gK (x) = α + x0 β + 1/2x0 Γx so the regression model to fit is y = α + x0 β + 1/2x0 Γx + ε • While the regression model has enough free parameters to be Diewert-flexible, the question remains: is plimαˆ = g(0)? Is plimβˆ = Dx g(0)? Is plimΓˆ = D2x g(0)? • The answer is no, in general. The reason is that if we treat the true values of the parameters as these derivatives, then ε is forced to play the part of the

remainder term, which is a function of x, so that x and ε are correlated in this case. As before, the estimator is biased in this case.


183

• A simpler example would be to consider a first-order T.S. approximation to a quadratic function. Draw picture. • The conclusion is that “flexible functional forms” aren’t really flexible in a useful statistical sense, in that neither the function itself nor its derivatives are consistently estimated, unless the function belongs to the parametric family of the specified functional form. In order to lead to consistent inferences, the regression model must be correctly specified.

10.1.1. The translog form. In spite of the fact that FFF’s aren’t really flexible for the purposes of econometric estimation and inference, they are useful, and they are certainly subject to less bias due to misspecification of the functional form than are many popular forms, such as the Cobb-Douglas or the simple linear in the variables model. The translog model is probably the most widely used FFF. This model is as above, except that the variables are subjected to a logarithmic tranformation. Also, the expansion point is usually taken to be the sample mean of the data, after the logarithmic transformation. The model is defined by y = ln(c) z x = ln z¯

= ln(z) − ln(¯z)

y = α + x0 β + 1/2x0 Γx + ε


184

In this presentation, the t subscript that distinguishes observations is suppressed for simplicity. Note that ∂y = β + Γx ∂x ∂ ln(c) (the other part of x is constant) = ∂ ln(z) =

∂c z ∂z c

which is the elasticity of c with respect to z. This is a convenient feature of the translog model. Note that at the means of the conditioning variables, z¯, x = 0, so ∂y =β ∂x z=¯z

so the β are the first-order elasticities, at the means of the data. To illustrate, consider that y is cost of production: y = c(w, q) where w is a vector of input prices and q is output. We could add other variables by extending q in the obvious manner, but this is supressed for simplicity. By Shephard’s lemma, the conditional factor demands are x=

∂c(w, q) ∂w

and the cost shares of the factors are therefore s=

wx ∂c(w, q) w = c ∂w c


185

which is simply the vector of elasticities of cost with respect to input prices. If the cost function is modeled using a translog function, we have h

i



ln(c) = α + x0 β + z0 δ + 1/2 x0 z 

Γ11 Γ12 Γ012

Γ22

 

x z

= α + x0 β + z0 δ + 1/2x0 Γ11 x + x0 Γ12 z + 1/2z2 γ22

 

where x = ln(w/w) ¯ (element-by-element division) and z = ln(q/q), ¯ and 

Γ11 = 

γ11 γ12

γ12 γ22   γ13  =  γ23

Γ12

 

Γ22 = γ33 .

Note that symmetry of the second derivatives has been imposed. Then the share equations are just s = β+

h

Γ11 Γ12

i

 

x z

 

Therefore, the share equations and the cost equation have parameters in common. By pooling the equations together and imposing the (true) restriction that the parameters of the equations be the same, we can gain efficiency. To illustrate in more detail, consider the case of two inputs, so 

x=

x1 x2



.


186

In this case the translog model of the logarithmic cost function is ln c = α + β1 x1 + β2 x2 + δz +

γ11 2 γ22 2 γ33 2 x + x + z + γ12 x1 x2 + γ13 x1 z + γ23 x2 z 2 1 2 2 2

The two cost shares of the inputs are the derivatives of ln c with respect to x 1 and x2 : s1 = β1 + γ11 x1 + γ12 x2 + γ13 z s2 = β2 + γ12 x1 + γ22 x2 + γ13 z Note that the share equations and the cost equation have parameters in common. One can do a pooled estimation of the three equations at once, imposing that the parameters are the same. In this way we’re using more observations and therefore more information, which will lead to imporved efficiency. Note that this does assume that the cost equation is correctly specified (i.e., not an approximation), since otherwise the derivatives would not be the true derivatives of the log cost function, and would then be misspecified for the shares. To pool the equations, write the model in matrix form (adding in error terms) 



ln c

   s1  s2





    = 0 1   0 0

x21 2

x22 2

z2 2

x1 x2

0

0 x1 0

0

x2

1

0 0

x2 0

x1

1 x 1 x2 z

          x1 z x 2 z    z 0     0 z          

α β1 β2 δ γ11 γ22 γ33 γ12 γ13 γ23



             ε 1      + ε    2     ε3          


187

This is one observation on the three equations. With the appropriate notation, a single observation can be written as yt = Xt θ + εt The overall model would stack n observations on the three equations for a total of 3n observations:

        





y1      y2    =  ..   .      yn





X1      X2    θ +   ..   .     Xn



ε1   ε2   ..  .    εn

Next we need to consider the errors. For observation t the errors can be placed in a vector



ε1t

  εt =  ε2t  ε3t

    

First consider the covariance matrix of this vector: the shares are certainly correlated since they must sum to one. (In fact, with 2 shares the variances are equal and the covariance is -1 times the variance. General notation is used to allow easy extension to the case of more than 2 inputs). Also, it’s likely that the shares and the cost equation have different variances. Supposing that the model is covariance stationary, the variance of εt won0 t depend upon t: 

σ11 σ12 σ13

  Varεt = Σ0 =  ·  ·

σ22 σ23 ·

σ33

    

Note that this matrix is singular, since the shares sum to 1. Assuming that there is no autocorrelation, the overall covariance matrix has the seemingly unrelated regressions


188

(SUR) structure. 

    Var    



ε1   ε2   = Σ ..  .    εn      =    

Σ0 0

··· .. .

0 .. .

Σ0 .. .

0

··· 0

= In ⊗ Σ0

0 .. .



     0    Σ0

where the symbol ⊗ indicates the Kronecker product. The Kronecker product of two matrices A and B is 

    A⊗B =    

a11 B a12 B · · · . a21 B . . .. .

a1q B .. .

a pq B · · ·

a pq B



    .   

10.1.2. FGLS estimation of a translog model. So, this model has heteroscedasticity and autocorrelation, so OLS won’t be efficient. The next question is: how do we estimate efficiently using FGLS? FGLS is based upon inverting the estimated error ˆ So we need to estimate Σ. covariance Σ. An asymptotically efficient procedure is (supposing normality of the errors)

(1) Estimate each equation by OLS


(2) Estimate Σ0 using

189

1 n Σˆ 0 = ∑ εˆ t εˆ t0 n t=1

(3) Next we need to account for the singularity of Σ0 . It can be shown that Σˆ 0 will be singular when the shares sum to one, so FGLS won’t work. The solution is to drop one of the share equations, for example the second. The model becomes 

 

ln c s1





=

x22

z2 2

x1 x2

0 x1 0

0

x2

1 x 1 x2 z 0 1

0

x21 2

2

or in matrix notation for the observation:

            x1 z x 2 z     z 0           

α β1 β2 δ γ11 γ22 γ33 γ12 γ13 γ23



               ε1 +    ε2           

yt∗ = Xt∗ θ + εt∗ and in stacked notation for all observations we have the 2n observations:       ∗ ∗ ∗  ε1   y1   X1         ε∗   y∗   X ∗   2   2   2   .  =  . θ+ .   ..   ..   ..              ∗ ∗ ∗ εn Xn yn


190

or, finally in matrix notation for all observations: y∗ = X ∗ θ + ε ∗ Considering the error covariance, we can define 

Σ∗0 = Var 

ε1 ε2

Σ∗ = In ⊗ Σ∗0

 

Define Σˆ ∗0 as the leading 2 × 2 block of Σˆ 0 , and form Σˆ ∗ = In ⊗ Σˆ ∗0 . This is a consistent estimator, following the consistency of OLS and applying a LLN. (4) Next compute the Cholesky factorization Pˆ0 = Chol Σˆ ∗0

−1

(I am assuming this is defined as an upper triangular matrix, which is consistent with the way Octave does it) and the Cholesky factorization of the overall covariance matrix of the 2 equation model, which can be calculated as Pˆ = Chol Σˆ ∗ = In ⊗ Pˆ0 (5) Finally the FGLS estimator can be calculated by applying OLS to the transformed model ˆ0 Pˆ 0 y∗ = Pˆ 0 X ∗ θ + ˆPε∗


191

or by directly using the GLS formula −1 −1 ∗ ˆθFGLS = X ∗0 Σˆ ∗0 −1 X ∗ X ∗0 Σˆ ∗0 y

It is equivalent to transform each observation individually: Pˆ00 y∗y = Pˆ00 Xt∗ θ + Pˆ00 ε∗ and then apply OLS. This is probably the simplest approach. A few last comments. (1) We have assumed no autocorrelation across time. This is clearly restrictive. It is relatively simple to relax this, but we won’t go into it here. (2) Also, we have only imposed symmetry of the second derivatives. Another restriction that the model should satisfy is that the estimated shares should sum to 1. This can be accomplished by imposing β1 + β 2 = 1 3

∑ γi j

= 0, j = 1, 2, 3.

i=1

These are linear parameter restrictions, so they are easy to impose and will improve efficiency if they are true. (3) The estimation procedure outlined above can be iterated. That is, estimate θˆ FGLS as above, then re-estimate Σ∗0 using errors calculated as εˆ = y − X θˆ FGLS These might be expected to lead to a better estimate than the estimator based on θˆ OLS , since FGLS is asymptotically more efficient. Then re-estimate θ using the new estimated error covariance. It can be shown that if this is

10.2. TESTING NONNESTED HYPOTHESES

192

repeated until the estimates don’t change (i.e., iterated to convergence) then the resulting estimator is the MLE. At any rate, the asymptotic properties of the iterated and uniterated estimators are the same, since both are based upon a consistent estimator of the error covariance.

10.2. Testing nonnested hypotheses Given that the choice of functional form isn’t perfectly clear, in that many possibilities exist, how can one choose between forms? When one form is a parametric restriction of another, the previously studied tests such as Wald, LR, score or qF are all possibilities. For example, the Cobb-Douglas model is a parametric restriction of the translog: The translog is yt = α + xt0 β + 1/2xt0 Γxt + ε where the variables are in logarithms, while the Cobb-Douglas is yt = α + xt0 β + ε so a test of the Cobb-Douglas versus the translog is simply a test that Γ = 0. The situation is more complicated when we want to test non-nested hypotheses. If the two functional forms are linear in the parameters, and use the same transformation of the dependent variable, then they may be written as M1 : y = X β + ε εt ∼ iid(0, σ2ε ) M2 : y = Zγ + η η ∼ iid(0, σ2η )


193

We wish to test hypotheses of the form: H0 : Mi is correctly specified versus HA : Mi is misspecified, for i = 1, 2. • One could account for non-iid errors, but we’ll suppress this for simplicity. • There are a number of ways to proceed. We’ll consider the J test, proposed by Davidson and MacKinnon, Econometrica (1981). The idea is to artificially nest the two models, e.g., y = (1 − α)X β + α(Zγ) + ω If the first model is correctly specified, then the true value of α is zero. On the other hand, if the second model is correctly specified then α = 1. – The problem is that this model is not identified in general. For example, if the models share some regressors, as in

M1 : yt = β1 + β2 x2t + β3 x3t + εt M2 : yt = γ1 + γ2 x2t + γ3 x4t + ηt then the composite model is yt = (1 − α)β1 + (1 − α)β2 x2t + (1 − α)β3 x3t + αγ1 + αγ2 x2t + αγ3 x4t + ωt Combining terms we get yt = ((1 − α)β1 + αγ1 ) + ((1 − α)β2 + αγ2 ) x2t + (1 − α)β3 x3t + αγ3 x4t + ωt = δ1 + δ2 x2t + δ3 x3t + δ4 x4t + ωt The four δ0 s are consistently estimable, but α is not, since we have four equations in 7 unknowns, so one can’t test the hypothesis that α = 0.


194

The idea of the J test is to substitute γˆ in place of γ. This is a consistent estimator supposing that the second model is correctly specified. It will tend to a finite probability limit even if the second model is misspecified. Then estimate the model y = (1 − α)X β + α(Zˆγ) + ω = X θ + αyˆ + ω where yˆ = Z(Z 0 Z)−1 Z 0 y = PZ y. In this model, α is consistently estimable, and one p

can show that, under the hypothesis that the first model is correct, α → 0 and that the ordinary t -statistic for α = 0 is asymptotically normal: t=

αˆ a ∼ N(0, 1) σˆ αˆ p

• If the second model is correctly specified, then t → ∞, since αˆ tends in probability to 1, while it’s estimated standard error tends to zero. Thus the test will always reject the false null model, asymptotically, since the statistic will eventually exceed any critical value with probability one. • We can reverse the roles of the models, testing the second against the first. • It may be the case that neither model is correctly specified. In this case, the test will still reject the null hypothesis, asymptotically, if we use critical values from the N(0, 1) distribution, since as long as αˆ tends to something p

different from zero, |t| → ∞. Of course, when we switch the roles of the models the other will also be rejected asymptotically. • In summary, there are 4 possible outcomes when we test two models, each against the other. Both may be rejected, neither may be rejected, or one of the two may be rejected.


195

• There are other tests available for non-nested models. The J− test is simple to apply when both models are linear in the parameters. The P-test is similar, but easier to apply when M1 is nonlinear. • The above presentation assumes that the same transformation of the dependent variable is used by both models. MacKinnon, White and Davidson, Journal of Econometrics, (1983) shows how to deal with the case of different transformations. • Monte-Carlo evidence shows that these tests often over-reject a correctly specified model. Can use bootstrap critical values to get better-performing tests.

CHAPTER 11

Exogeneity and simultaneity Several times we’ve encountered cases where correlation between regressors and the error term lead to biasedness and inconsistency of the OLS estimator. Cases include autocorrelation with lagged dependent variables and measurement error in the regressors. Another important case is that of simultaneous equations. The cause is different, but the effect is the same.

11.1. Simultaneous equations Up until now our model is y = Xβ + ε where, for purposes of estimation we can treat X as fixed. This means that when estimating β we condition on X . When analyzing dynamic models, we’re not interested in conditioning on X , as we saw in the section on stochastic regressors. Nevertheless, the OLS estimator obtained by treating X as fixed continues to have desirable asymptotic properties even in that case. 196

11.1. SIMULTANEOUS EQUATIONS

197

Simultaneous equations is a different prospect. An example of a simultaneous equation system is a simple supply-demand system: Demand: qt = α1 + α2 pt + α3 yt + ε1t 

E 

ε1t ε2t

Supply: qt = β1 + β2 pt + ε2t    h i σ σ  ε ε  =  11 12  1t 2t · σ22 

≡ Σ, ∀t

The presumption is that qt and pt are jointly determined at the same time by the intersection of these equations. We’ll assume that yt is determined by some unrelated process. It’s easy to see that we have correlation between regressors and errors. Solving for pt : α1 + α2 pt + α3 yt + ε1t = β1 + β2 pt + ε2t β2 pt − α2 pt = α1 − β1 + α3 yt + ε1t − ε2t pt =

α1 − β 1 α3 yt ε1t − ε2t + + β2 − α 2 β2 − α 2 β2 − α 2

Now consider whether pt is uncorrelated with ε1t :

α3 yt ε1t − ε2t α1 − β 1 E (pt ε1t ) = E + + β2 − α 2 β2 − α 2 β2 − α 2 σ11 − σ12 = β2 − α 2

ε1t

Because of this correlation, OLS estimation of the demand equation will be biased and inconsistent. The same applies to the supply equation, for the same reason. In this model, qt and pt are the endogenous varibles (endogs), that are determined within the system. yt is an exogenous variable (exogs). These concepts are a bit tricky,

11.1. SIMULTANEOUS EQUATIONS

198

and we’ll return to it in a minute. First, some notation. Suppose we group together current endogs in the vector Yt . If there are G endogs, Yt is G × 1. Group current and lagged exogs, as well as lagged endogs in the vector Xt , which is K ×1. Stack the errors of the G equations into the error vector Et . The model, with additional assumtions, can be written as Yt0 Γ = Xt0 B + Et0 Et ∼ N(0, Σ), ∀t E (Et Es0 ) = 0,t 6= s

We can stack all n observations and write the model as Y Γ = XB + E E (X 0 E) = 0(K×G)

vec(E) ∼ N(0, Ψ) where



    Y =   

Y10





X10

     X0   2  ,X =  . ..   ..  .     0 Yn Xn0 Y20

Y is n × G, X is n × K, and E is n × G.





        ,E =       

E10 E20

.. .

En0

        

• This system is complete, in that there are as many equations as endogs. • There is a normality assumption. This isn’t necessary, but allows us to consider the relationship between least squares and ML estimators.

11.2. EXOGENEITY

199

• Since there is no autocorrelation of the Et ’s, and since the columns of E are individually homoscedastic, then   σ11 In σ12 In · · · σ1G In  ..  σ22 In .  Ψ =  . . ..  . .   · σGG In = In ⊗ Σ

        

• X may contain lagged endogenous and exogenous variables. These variables are predetermined. • We need to define what is meant by “endogenous” and “exogenous” when classifying the current period variables.

11.2. Exogeneity The model defines a data generating process. The model involves two sets of variables, Yt and Xt , as well as a parameter vector θ=

h

vec(Γ)0

vec(B)0

vec∗ (Σ)0

i0

• In general, without additional restrictions, θ is a G2 + GK + G2 − G /2 + G dimensional vector. This is the parameter vector that were interested in

estimating. • In principle, there exists a joint density function for Yt and Xt , which depends on a parameter vector φ. Write this density as ft (Yt , Xt |φ, It )

11.2. EXOGENEITY

200

where It is the information set in period t. This includes lagged Yt0 s and lagged Xt ’s of course. This can be factored into the density of Yt conditional on Xt times the marginal density of Xt : ft (Yt , Xt |φ, It ) = ft (Yt |Xt , φ, It ) ft (Xt |φ, It ) This is a general factorization, but is may very well be the case that not all parameters in φ affect both factors. So use φ1 to indicate elements of φ that enter into the conditional density and write φ2 for parameters that enter into the marginal. In general, φ1 and φ2 may share elements, of course. We have ft (Yt , Xt |φ, It ) = ft (Yt |Xt , φ1 , It ) ft (Xt |φ2 , It ) • Recall that the model is Yt0 Γ = Xt0 B + Et0 Et ∼ N(0, Σ), ∀t E (Et Es0 ) = 0,t 6= s

Normality and lack of correlation over time imply that the observations are independent of one another, so we can write the log-likelihood function as the sum of likelihood contributions of each observation: ln L(Y |θ, It ) = = =

n

∑ ln ft (Yt , Xt |φ, It )

t=1 n

∑ ln ( ft (Yt |Xt , φ1, It ) ft (Xt |φ2, It ))

t=1 n

n

t=1

t=1

∑ ln ft (Yt |Xt , φ1, It ) + ∑ ln ft (Xt |φ2, It ) =

11.2. EXOGENEITY

201

D EFINITION 15 (Weak Exogeneity). Xt is weakly exogeneous for θ (the original parameter vector) if there is a mapping from φ to θ that is invariant to φ 2 . More formally, for an arbitrary (φ1 , φ2 ), θ(φ) = θ(φ1 ). This implies that φ1 and φ2 cannot share elements if Xt is weakly exogenous, since φ1 would change as φ2 changes, which prevents consideration of arbitrary combinations of (φ1 , φ2 ). Supposing that Xt is weakly exogenous, then the MLE of φ1 using the joint density is the same as the MLE using only the conditional density ln L(Y |X , θ, It ) =

n

∑ ln ft (Yt |Xt , φ1, It )

t=1

since the conditional likelihood doesn’t depend on φ2 . In other words, the joint and conditional log-likelihoods maximize at the same value of φ1 . • With weak exogeneity, knowledge of the DGP of Xt is irrelevant for inference

on φ1 , and knowledge of φ1 is sufficient to recover the parameter of interest, θ. Since the DGP of Xt is irrelevant, we can treat Xt as fixed in inference.

• By the invariance property of MLE, the MLE of θ is θ(φˆ 1 ),and this mapping is assumed to exist in the definition of weak exogeneity. • Of course, we’ll need to figure out just what this mapping is to recover θˆ from φˆ 1 . This is the famous identification problem.

• With lack of weak exogeneity, the joint and conditional likelihood functions maximize in different places. For this reason, we can’t treat Xt as fixed in inference. The joint MLE is valid, but the conditional MLE is not. • In resume, we require the variables in Xt to be weakly exogenous if we are to be able to treat them as fixed in estimation. Lagged Yt satisfy the definition, since they are in the conditioning information set, e.g., Yt−1 ∈ It . Lagged

11.3. REDUCED FORM

202

Yt aren’t exogenous in the normal usage of the word, since their values are determined within the model, just earlier on. Weakly exogenous variables include exogenous (in the normal sense) variables as well as all predetermined variables. 11.3. Reduced form Recall that the model is

Yt0 Γ = Xt0 B + Et0 V (Et ) = Σ This is the model in structural form. D EFINITION 16 (Structural form). An equation is in structural form when more than one current period endogenous variable is included. The solution for the current period endogs is easy to find. It is Yt0 = Xt0 BΓ−1 + Et0 Γ−1 = Xt0 Π +Vt0 = Now only one current period endog appears in each equation. This is the reduced form. D EFINITION 17 (Reduced form). An equation is in reduced form if only one current period endog is included. An example is our supply/demand system. The reduced form for quantity is obtained by solving the supply equation for price and substituting into demand:

11.3. REDUCED FORM

qt = α1 + α2

203

qt − β1 − ε2t β2

+ α3 yt + ε1t

β2 qt − α2 qt = β2 α1 − α2 (β1 + ε2t ) + β2 α3 yt + β2 ε1t

β2 ε1t − α2 ε2t β2 α1 − α2 β1 β2 α3 yt + + β2 − α 2 β2 − α 2 β2 − α 2

qt =

= π11 + π21 yt +V1t Similarly, the rf for price is

β1 + β2 pt + ε2t = α1 + α2 pt + α3 yt + ε1t β2 pt − α2 pt = α1 − β1 + α3 yt + ε1t − ε2t

α1 − β 1 α3 yt ε1t − ε2t + + β2 − α 2 β2 − α 2 β2 − α 2

pt =

= π12 + π22 yt +V2t

The interesting thing about the rf is that the equations individually satisfy the classical assumptions, since yt is uncorrelated with ε1t and ε2t by assumption, and therefore E (yt Vit ) = 0, i=1,2, ∀t. The errors of the rf are



V1t



V2t

V (V1t ) = E

The variance of V1t is

=





=

β2 ε1t −α2 ε2t β2 −α2 ε1t −ε2t β2 −α2

β2 ε1t − α2 ε2t β2 − α 2

 

β2 ε1t − α2 ε2t β2 − α 2

β22 σ11 − 2β2 α2 σ12 + α2 σ22

(β2 − α2 )2

• This is constant over time, so the first rf equation is homoscedastic.

• Likewise, since the εt are independent over time, so are the Vt .

11.4. IV ESTIMATION

204

The variance of the second rf error is

ε1t − ε2t ε1t − ε2t V (V2t ) = E β2 − α 2 β2 − α 2 σ11 − 2σ12 + σ22 = (β2 − α2 )2 and the contemporaneous covariance of the errors across equations is E (V1t V2t ) = E

=

β2 ε1t − α2 ε2t β2 − α 2

ε1t − ε2t β2 − α 2

β2 σ11 − (β2 + α2 ) σ12 + σ22

(β2 − α2 )2

• In summary the rf equations individually satisfy the classical assumptions, under the assumtions we’ve made, but they are contemporaneously correlated. The general form of the rf is Yt0 = Xt0 BΓ−1 + Et0 Γ−1 = Xt0 Π +Vt0 so we have that

0 0 Vt = Γ−1 Et ∼ N 0, Γ−1 ΣΓ−1 , ∀t

and that the Vt are timewise independent (note that this wouldn’t be the case if the Et were autocorrelated).

11.4. IV estimation The IV estimator may appear a bit unusual at first, but it will grow on you over time.

11.4. IV ESTIMATION

205

The simultaneous equations model is Y Γ = XB + E Considering the first equation (this is without loss of generality, since we can always reorder the equations) we can partition the Y matrix as Y=

h

y Y1 Y2

i

• y is the first column • Y1 are the other endogenous variables that enter the first equation • Y2 are endogs that are excluded from this equation

Similarly, partition X as X=

h

X1 X2

i

• X1 are the included exogs, and X2 are the excluded exogs.

Finally, partition the error matrix as E=

h

ε E12

i

Assume that Γ has ones on the main diagonal. These are normalization restrictions that simply scale the remaining coefficients on each equation, and which scale the variances of the error terms.

11.4. IV ESTIMATION

206

Given this scaling and our partitioning, the coefficient matrices can be written as 



1 Γ12     Γ =  −γ1 Γ22    0 Γ32   β1 B12  B =  0 B22

With this, the first equation can be written as

y = Y1 γ1 + X1 β1 + ε = Zδ + ε The problem, as we’ve seen is that Z is correlated with ε, since Y1 is formed of endogs. Now, let’s consider the general problem of a linear regression model with correlation between regressors and the error term: y = Xβ + ε ε ∼ iid(0, In σ2 ) E (X 0 ε) 6= 0.

The present case of a structural equation from a system of equations fits into this notation, but so do other problems, such as measurement error or lagged dependent variables with autocorrelated errors. Consider some matrix W which is formed of variables uncorrelated with ε. This matrix defines a projection matrix PW = W (W 0W )−1W 0

11.4. IV ESTIMATION

207

so that anything that is projected onto the space spanned by W will be uncorrelated with ε, by the definition of W. Transforming the model with this projection matrix we get PW y = PW X β + PW ε or y∗ = X ∗ β + ε ∗ Now we have that ε∗ and X ∗ are uncorrelated, since this is simply 0 E (X ∗0 ε∗ ) = E (X 0 PW PW ε)

= E (X 0 PW ε) and PW X = W (W 0W )−1W 0 X is the fitted value from a regression of X on W. This is a linear combination of the columns of W, so it must be uncorrelated with ε. This implies that applying OLS to the model y∗ = X ∗ β + ε ∗ will lead to a consistent estimator, given a few more assumptions. This is the generalized instrumental variables estimator. W is known as the matrix of instruments. The estimator is βˆ IV = (X 0 PW X )−1 X 0 PW y from which we obtain βˆ IV = (X 0 PW X )−1 X 0 PW (X β + ε) = β + (X 0 PW X )−1 X 0 PW ε

11.4. IV ESTIMATION

208

so βˆ IV − β = (X 0PW X )−1 X 0 PW ε X 0W (W 0W )−1W 0 X

=

Now we can introduce factors of n to get

βˆ IV − β =

X 0W n

W 0W −1 n

!

W 0X n

−1

X 0W (W 0W )−1W 0 ε

!−1

X 0W n

W 0W n

−1

W 0ε n

Assuming that each of the terms with a n in the denominator satisfies a LLN, so that • • •

W 0W n X 0W n

p

→ QWW , a finite pd matrix p

→ QXW , a finite matrix with rank K (= cols(X ) )

W 0ε p n →0

then the plim of the rhs is zero. This last term has plim 0 since we assume that W and ε are uncorrelated, e.g., E (Wt0 εt ) = 0,

Given these assumtions the IV estimator is consistent p βˆ IV → β.

Furthermore, scaling by √ ˆ n βIV − β =

√

n, we have

X 0W n

W 0W n

−1

W 0X n

!−1

X 0W n

W 0W n

Assuming that the far right term satifies a CLT, so that •

0 d W √ ε → N(0, QWW σ2 ) n

then we get

√ ˆ d −1 Q0XW )−1 σ2 n βIV − β → N 0, (QXW QWW

−1

W 0ε √ n

11.4. IV ESTIMATION

209

The estimators for QXW and QWW are the obvious ones. An estimator for σ2 is 0 1 d 2 ˆ ˆ σIV = y − X βIV y − X βIV . n

This estimator is consistent following the proof of consistency of the OLS estimator of σ2 , when the classical assumptions hold. The formula used to estimate the variance of βˆ IV is Vˆ (βˆ IV ) = The IV estimator is

X 0W

W 0W

−1

W 0X

−1 d σ2IV

(1) Consistent (2) Asymptotically normally distributed (3) Biased in general, since even though E (X 0 PW ε) = 0, E (X 0 PW X )−1 X 0 PW ε may not be zero, since (X 0 PW X )−1 and X 0 PW ε are not independent. An important point is that the asymptotic distribution of βˆ IV depends upon QXW and QWW , and these depend upon the choice of W. The choice of instruments influences the efficiency of the estimator. • When we have two sets of instruments, W1 and W2 such that W1 ⊂ W2 , then the IV estimator using W2 is at least as efficiently asymptotically as the estimator that used W1 . More instruments leads to more asymptotically efficient estimation, in general. • There are special cases where there is no gain (simultaneous equations is an example of this, as we’ll see). • The penalty for indiscriminant use of instruments is that the small sample bias of the IV estimator rises as the number of instruments increases. The reason

11.5. IDENTIFICATION BY EXCLUSION RESTRICTIONS

210

for this is that PW X becomes closer and closer to X itself as the number of instruments increases. • IV estimation can clearly be used in the case of simultaneous equations. The only issue is which instruments to use.

11.5. Identification by exclusion restrictions The identification problem in simultaneous equations is in fact of the same nature as the identification problem in any estimation setting: does the limiting objective function have the proper curvature so that there is a unique global minimum or maximum at the true parameter value? In the context of IV estimation, this is the case if the limiting covariance of the IV estimator is positive definite and plim 1n W 0 ε = 0. This matrix is −1 V∞ (βˆ IV ) = (QXW QWW Q0XW )−1 σ2

• The necessary and sufficient condition for identification is simply that this matrix be positive definite, and that the instruments be (asymptotically) uncorrelated with ε. • For this matrix to be positive definite, we need that the conditions noted above hold: QWW must be positive definite and QXW must be of full rank ( K ). • These identification conditions are not that intuitive nor is it very obvious how to check them.

11.5.1. Necessary conditions. If we use IV estimation for a single equation of the system, the equation can be written as y = Zδ + ε


where Z= Notation:

h

Y1 X1

211

i

• Let K be the total numer of weakly exogenous variables. • Let K ∗ = cols(X1 ) be the number of included exogs, and let K ∗∗ = K − K ∗ be the number of excluded exogs (in this equation). • Let G∗ = cols(Y1 ) + 1 be the total number of included endogs, and let G∗∗ = G − G∗ be the number of excluded endogs. Using this notation, consider the selection of instruments. • Now the X1 are weakly exogenous and can serve as their own instruments. • It turns out that X exhausts the set of possible instruments, in that if the variables in X don’t lead to an identified model then no other instruments will identify the model either. Assuming this is true (we’ll prove it in a moment), then a necessary condition for identification is that cols(X2 ) ≥ cols(Y1 ) since if not then at least one instrument must be used twice, so W will not have full column rank: ρ(W ) < K ∗ + G∗ − 1 ⇒ ρ(QZW ) < K ∗ + G∗ − 1 This is the order condition for identification in a set of simultaneous equations. When the only identifying information is exclusion restrictions on the variables that enter an equation, then the number of excluded exogs must be greater than or equal to the number of included endogs, minus 1 (the normalized lhs endog), e.g., K ∗∗ ≥ G∗ − 1


212

• To show that this is in fact a necessary condition consider some arbitrary set of instruments W. A necessary condition for identification is that

1 ρ plim W 0 Z n where Z=

h

= K ∗ + G∗ − 1

Y1 X1

Recall that we’ve partitioned the model

i

Y Γ = XB + E as Y=

Given the reduced form

h

y Y1 Y2 h i X = X1 X2

i

Y = X Π +V we can write the reduced form using the same partition h so we have

y Y1 Y2

i

=

h

X1 X2

i

 

π11 Π12 Π13 π21 Π22 Π23



+

h

v V1 V2

Y1 = X1 Π12 + X2 Π22 +V1 so

i 1 0 1 h W Z = W 0 X1 Π12 + X2 Π22 +V1 X1 n n

i


213

Because the W ’s are uncorrelated with the V1 ’s, by assumption, the cross between W and V1 converges in probability to zero, so i 1 1 h plim W 0 Z = plim W 0 X1 Π12 + X2 Π22 X1 n n Since the far rhs term is formed only of linear combinations of columns of X , the rank of this matrix can never be greater than K, regardless of the choice of instruments. If Z has more than K columns, then it is not of full column rank. When Z has more than K columns we have G∗ − 1 + K ∗ > K or noting that K ∗∗ = K − K ∗ ,

G∗ − 1 > K ∗∗

In this case, the limiting matrix is not of full column rank, and the identification condition fails.

11.5.2. Sufficient conditions. Identification essentially requires that the structural parameters be recoverable from the data. This won’t be the case, in general, unless the structural model is subject to some restrictions. We’ve already identified necessary conditions. Turning to sufficient conditions (again, we’re only considering identification through zero restricitions on the parameters, for the moment). The model is Yt0 Γ = Xt0 B + Et V (Et ) = Σ


214

This leads to the reduced form Yt0 = Xt0 BΓ−1 + Et Γ−1 = Xt0 Π +Vt V (Vt ) =

0 Γ−1 ΣΓ−1

= Ω

The reduced form parameters are consistently estimable, but none of them are known a priori, and there are no restrictions on their values. The problem is that more than one structural form has the same reduced form, so knowledge of the reduced form parameters alone isn’t enough to determine the structural parameters. To see this, consider the model Yt0 ΓF = Xt0 BF + Et F V (Et F) = F 0 ΣF where F is some arbirary nonsingular G × G matrix. The rf of this new model is Yt0 = Xt0 BF (ΓF)−1 + Et F (ΓF)−1 = Xt0 BFF −1 Γ−1 + Et FF −1 Γ−1 = Xt0 BΓ−1 + Et Γ−1 = Xt0 Π +Vt Likewise, the covariance of the rf of the transformed model is V (Et F (ΓF)−1 ) = V (Et Γ−1 ) = Ω


215

Since the two structural forms lead to the same rf, and the rf is all that is directly estimable, the models are said to be observationally equivalent. What we need for identification are restrictions on Γ and B such that the only admissible F is an identity matrix (if all of the equations are to be identified). Take the coefficient matrices as partitioned before: 

       Γ  =   B   

1 −γ1 0 β1 0



Γ12   Γ22    Γ32     B12   B22

The coefficients of the first equation of the transformed model are simply these coefficients multiplied by the first column of F. This gives  Γ12  1      −γ1 Γ22   Γ f11   = 0 Γ32   B F2  B12  β1  0 B22



      f11     F2   

For identification of the first equation we need that there be enough restrictions so that

the only admissible

 

f11 F2

 


be the leading column of an identity matrix, so that    Γ12    1     −γ1 Γ22          f11     0   = Γ 32      F2     B12    β1    0 B22

Note that the third and fifth rows are     Γ 0  32  F2 =   B22 0

1

216



  −γ1     0    β1   0

Supposing that the leading matrix is of full column rank, e.g., 

ρ 

Γ32 B22





 = cols 

Γ32 B22



 = G − 1

then the only way this can hold, without additional restrictions on the model’s parameters, is if F2 is a vector of zeros. Given that F2 is a vector of zeros, then the first equation h Therefore, as long as

then

1 Γ12

i



 

ρ   

f11 F2

f11 F2

Γ32 B22 



 = 1 ⇒ f11 = 1



 = G − 1 

=

1 0G−1

 


217

The first equation is identified in this case, so the condition is sufficient for identification. It is also necessary, since the condition implies that this submatrix must have at least G − 1 rows. Since this matrix has G∗∗ + K ∗∗ = G − G∗ + K ∗∗ rows, we obtain G − G∗ + K ∗∗ ≥ G − 1 or K ∗∗ ≥ G∗ − 1 which is the previously derived necessary condition. The above result is fairly intuitive (draw picture here). The necessary condition ensures that there are enough variables not in the equation of interest to potentially move the other equations, so as to trace out the equation of interest. The sufficient condition ensures that those other equations in fact do move around as the variables change their values. Some points: • When an equation has K ∗∗ = G∗ − 1, is is exactly identified, in that omission of an identifiying restriction is not possible without loosing consistency. • When K ∗∗ > G∗ − 1, the equation is overidentified, since one could drop a restriction and still retain consistency. Overidentifying restrictions are therefore testable. When an equation is overidentified we have more instruments than are strictly necessary for consistent estimation. Since estimation by IV with more instruments is more efficient asymptotically, one should employ overidentifying restrictions if one is confident that they’re true. • We can repeat this partition for each equation in the system, to see which equations are identified and which aren’t.


218

• These results are valid assuming that the only identifying information comes from knowing which variables appear in which equations, e.g., by exclusion restrictions, and through the use of a normalization. There are other sorts of identifying information that can be used. These include (1) Cross equation restrictions (2) Additional restrictions on parameters within equations (as in the Klein model discussed below) (3) Restrictions on the covariance matrix of the errors (4) Nonlinearities in variables • When these sorts of information are available, the above conditions aren’t necessary for identification, though they are of course still sufficient.

To give an example of how other information can be used, consider the model Y Γ = XB + E where Γ is an upper triangular matrix with 1’s on the main diagonal. This is a triangular system of equations. In this case, the first equation is y1 = X B·1 + E·1 Since only exogs appear on the rhs, this equation is identified. The second equation is

y2 = −γ21 y1 + X B·2 + E·2 This equation has K ∗∗ = 0 excluded exogs, and G∗ = 2 included endogs, so it fails the order (necessary) condition for identification.


219

• However, suppose that we have the restriction Σ21 = 0, so that the first and second structural errors are uncorrelated. In this case

E (y1t ε2t ) = E (Xt0B·1 + ε1t )ε2t = 0

so there’s no problem of simultaneity. If the entire Σ matrix is diagonal, then following the same logic, all of the equations are identified. This is known as a fully recursive model.

11.5.3. Example: Klein’s Model 1. To give an example of determining identification status, consider the following macro model (this is the widely known Klein’s Model 1) p

g

Consumption: Ct = α0 + α1 Pt + α2 Pt−1 + α3 (Wt +Wt ) + ε1t Investment: It = β0 + β1 Pt + β2 Pt−1 + β3 Kt−1 + ε2t Private Wages: Wt

p

= γ0 + γ1 Xt + γ2 Xt−1 + γ3 At + ε3t

Output: Xt = Ct + It + Gt Profits: Pt = Xt − Tt −Wt Capital Stock:  ε1t    ε2t  ε3t

Kt = Kt−1 + It  

p

 

σ11 σ12 σ13 0          ∼ IID  0  ,  σ22 σ23     σ33 0 g

    

The other variables are the government wage bill, Wt , taxes, Tt , government nonwage spending, Gt ,and a time trend, At . The endogenous variables are the lhs variables, Yt0

=

h

Ct It Wt

p

Xt Pt Kt

i


220

and the predetermined variables are all others: Xt0

=

h

1

g Wt

Gt Tt At Pt−1 Kt−1 Xt−1

i

.

The model assumes that the errors of the equations are contemporaneously correlated, by nonautocorrelated. The model written as Y Γ = X B + E gives 

       Γ=      

1

0

0

0

1

0

−α3 0

1

0

−γ1

0

−α1 −β1 0 0

0 

          B=          

0

−1 0

  −1 0 −1    0 1 0    1 −1 0     0 1 0   0 0 1

α 0 β0 γ 0 0 0 α3 0

0

0

0

0

0

0

0

0

0

γ3

α 2 β2 0 0

β3 0

0

0

γ2

0





0   0 0 0    1 0 0     0 −1 0    0 0 0    0 0 0    0 0 1    0 0 0

To check this identification of the consumption equation, we need to extract Γ 32 and B22 , the submatrices of coefficients of endogs and exogs that don’t appear in this equation. These are the rows that have zeros in the first column, and we need to drop the


221

first column. We get 

 

Γ32 B22

           =          

1

0

0

−γ1 1

0

0

0

0

0

1

0

0

0

0

γ3

0

β3 0

0

γ2

0

0

−1 0



−1   −1 0    0 1     0 0    −1 0    0 0    0 1    0 0

We need to find a set of 5 rows of this matrix gives a full-rank 5×5 matrix. For example, selecting rows 3,4,5,6, and 7 we obtain the matrix   1   0 0 0 0     0 0 1 0 0      A =  0 0 0 −1 0       0 γ 0 0 0   3   β3 0 0 0 1

This matrix is of full rank, so the sufficient condition for identification is met. Counting included endogs, G∗ = 3, and counting excluded exogs, K ∗∗ = 5, so K ∗∗ − L = G∗ − 1 5−L L

= 3−1 =3

• The equation is over-identified by three restrictions, according to the counting rules, which are correct when the only identifying information are the

11.6. 2SLS

222

exclusion restrictions. However, there is additional information in this case. p

g

Both Wt and Wt enter the consumption equation, and their coefficients are restricted to be the same. For this reason the consumption equation is in fact overidentified by four restrictions.

11.6. 2SLS When we have no information regarding cross-equation restrictions or the structure of the error covariance matrix, one can estimate the parameters of a single equation of the system without regard to the other equations. • This isn’t always efficient, as we’ll see, but it has the advantage that misspecifications in other equations will not affect the consistency of the estimator of the parameters of the equation of interest. • Also, estimation of the equation won’t be affected by identification problems in other equations. The 2SLS estimator is very simple: in the first stage, each column of Y1 is regressed on all the weakly exogenous variables in the system, e.g., the entire X matrix. The fitted values are Yˆ1 = X (X 0X )−1 X 0Y1 = PX Y1 ˆ1 = XΠ Since these fitted values are the projection of Y1 on the space spanned by X , and since any vector in this space is uncorrelated with ε by assumption, Yˆ1 is uncorrelated with ε. Since Yˆ1 is simply the reduced-form prediction, it is correlated with Y1 , The only other requirement is that the instruments be linearly independent. This should be the

11.6. 2SLS

223

case when the order condition is satisfied, since there are more columns in X2 than in Y1 in this case. The second stage substitutes Yˆ1 in place of Y1 , and estimates by OLS. This original model is y = Y1 γ1 + X1 β1 + ε = Zδ + ε and the second stage model is y = Yˆ1 γ1 + X1 β1 + ε. Since X1 is in the space spanned by X , PX X1 = X1 , so we can write the second stage model as y = PX Y1 γ1 + PX X1 β1 + ε ≡ PX Zδ + ε The OLS estimator applied to this model is δˆ = (Z 0 PX Z)−1 Z 0 PX y which is exactly what we get if we estimate using IV, with the reduced form predictions of the endogs used as instruments. Note that if we define Zˆ = PX Z h i = Yˆ1 X1

11.6. 2SLS

224

so that Zˆ are the instruments for Z, then we can write δˆ = (Zˆ 0 Z)−1 Zˆ 0 y • Important note: OLS on the transformed model can be used to calculate the

2SLS estimate of δ, since we see that it’s equivalent to IV using a particular

set of instruments. However the OLS covariance formula is not valid. We need to apply the IV covariance formula already seen above.

Actually, there is also a simplification of the general IV variance formula. Define Zˆ = PX Z h i ˆ = Y X The IV covariance estimator would ordinarily be ˆ = Z 0 Zˆ Vˆ (δ)

−1

Zˆ 0 Zˆ

However, looking at the last term in brackets

Zˆ 0 Z =

h

Yˆ1 X1

i0 h

Y1 X1

i



=

Zˆ 0 Z

−1

σˆ 2IV

Y10 (PX )Y1

Y10 (PX )X1

X10 Y1

X10 X1

but since PX is idempotent and since PX X = X , we can write h

Yˆ1 X1

i0 h

Y1 X1

i



=  =

h

Y10 PX PX Y1

Y10 PX X1

X10 PX Y1

X10 X1

Yˆ1 X1

= Zˆ 0 Zˆ

i0 h

Yˆ1 X1

 

i

 

11.7. TESTING THE OVERIDENTIFYING RESTRICTIONS

225

Therefore, the second and last term in the variance formula cancel, so the 2SLS varcov estimator simplifies to ˆ = Z 0 Zˆ Vˆ (δ)

−1

σˆ 2IV

−1

σˆ 2IV

which, following some algebra similar to the above, can also be written as ˆ = Zˆ 0 Zˆ Vˆ (δ)

Finally, recall that though this is presented in terms of the first equation, it is general since any equation can be placed first. Properties of 2SLS: (1) Consistent (2) Asymptotically normal (3) Biased when the mean esists (the existence of moments is a technical issue we won’t go into here). (4) Asymptotically inefficient, except in special circumstances (more on this later).

11.7. Testing the overidentifying restrictions The selection of which variables are endogs and which are exogs is part of the specification of the model. As such, there is room for error here: one might erroneously classify a variable as exog when it is in fact correlated with the error term. A general test for the specification on the model can be formulated as follows: The IV estimator can be calculated by applying OLS to the transformed model, so the IV objective function at the minimized value is 0 s(βˆ IV ) = y − X βˆ IV PW y − X βˆ IV ,


226

but εˆ IV = y − X βˆ IV = y − X (X 0 PW X )−1 X 0 PW y = =

I − X (X 0PW X )−1 X 0 PW y

I − X (X 0PW X )−1 X 0 PW (X β + ε)

= A (X β + ε) where

A ≡ I − X (X 0PW X )−1 X 0 PW so s(βˆ IV ) = ε0 + β0 X 0 A0 PW A (X β + ε)

Moreover, A0 PW A is idempotent, as can be verified by multiplication: A0 PW A = = =

I − PW X (X 0PW X )−1 X 0 PW I − X (X 0PW X )−1 X 0 PW PW − PW X (X 0PW X )−1 X 0 PW

I − PW X (X 0PW X )−1 X 0 PW .

PW − PW X (X 0PW X )−1 X 0 PW

Furthermore, A is orthogonal to X AX =

I − X (X 0PW X )−1 X 0 PW X

= X −X = 0 so

s(βˆ IV ) = ε0 A0 PW Aε


227

Supposing the ε are normally distributed, with variance σ2 , then the random variable s(βˆ IV ) ε0 A0 PW Aε = σ2 σ2 is a quadratic form of a N(0, 1) random variable with an idempotent matrix in the middle, so

s(βˆ IV ) ∼ χ2 (ρ(A0 PW A)) σ2

This isn’t available, since we need to estimate σ2 . Substituting a consistent estimator, s(βˆ IV ) a 2 ∼ χ (ρ(A0 PW A)) c 2 σ

• Even if the ε aren’t normally distributed, the asymptotic result still holds. The last thing we need to determine is the rank of the idempotent matrix. We have A0 PW A = PW − PW X (X 0PW X )−1 X 0 PW so

ρ(A0PW A) = Tr PW − PW X (X 0PW X )−1 X 0 PW

= TrPW − TrX 0 PW PW X (X 0PW X )−1 = TrW (W 0W )−1W 0 − KX = TrW 0W (W 0W )−1 − KX = KW − KX

where KW is the number of columns of W and KX is the number of columns of X . The degrees of freedom of the test is simply the number of overidentifying restrictions: the number of instruments we have beyond the number that is strictly necessary for consistent estimation.


228

• This test is an overall specification test: the joint null hypothesis is that the model is correctly specified and that the W form valid instruments (e.g., that the variables classified as exogs really are uncorrelated with ε. Rejection can mean that either the model y = Zδ + ε is misspecified, or that there is correlation between X and ε. • This is a particular case of the GMM criterion test, which is covered in the second half of the course. See Section 15.8. • Note that since

εˆ IV = Aε

and s(βˆ IV ) = ε0 A0 PW Aε we can write s(βˆ IV ) = c2 σ

εˆ 0W (W 0W )−1W 0 W (W 0W )−1W 0 εˆ εˆ 0 εˆ /n

= n(RSSεˆ IV |W /T SSεˆ IV ) = nR2u

where R2u is the uncentered R2 from a regression of the IV residuals on all of the instruments W . This is a convenient way to calculate the test statistic.

On an aside, consider IV estimation of a just-identified model, using the standard notation

y = Xβ + ε


229

and W is the matrix of instruments. If we have exact identification then cols(W ) = 0

cols(X ), so W X is a square matrix. The transformed model is PW y = PW X β + PW ε and the fonc are X 0PW (y − X βˆ IV ) = 0 The IV estimator is βˆ IV = X 0 PW X Considering the inverse here X 0 PW X

−1

=

−1

X 0 PW y

X 0W (W 0W )−1W 0 X

−1

= (W 0 X )−1 X 0W (W 0W )−1 = (W 0 X )−1 (W 0W ) X 0W

Now multiplying this by X 0 PW y, we obtain βˆ IV = (W 0 X )−1 (W 0W ) X 0W = (W 0 X )−1 (W 0W ) X 0W = (W 0 X )−1W 0 y

−1

−1

−1

−1

X 0 PW y X 0W (W 0W )−1W 0 y

11.8. SYSTEM METHODS OF ESTIMATION

230

The objective function for the generalized IV estimator is

0 ˆ ˆ y − X βIV PW y − X βIV 0 0 0 ˆ ˆ ˆ y PW y − X βIV − βIV X PW y − X βIV y0 PW y − X βˆ IV − βˆ 0IV X 0 PW y + βˆ 0IV X 0 PW X βˆ IV y0 PW y − X βˆ IV − βˆ 0IV X 0 PW y + X 0 PW X βˆ IV y0 PW y − X βˆ IV

s(βˆ IV ) = = = = =

by the fonc for generalized IV. However, when we’re in the just indentified case, this is s(βˆ IV ) = y0 PW y − X (W 0 X )−1W 0 y

= y0 PW I − X (W 0 X )−1W 0 y

= y0 W (W 0W )−1W 0 −W (W 0W )−1W 0 X (W 0 X )−1W 0 y

= 0

The value of the objective function of the IV estimator is zero in the just identified case. This makes sense, since we’ve already shown that the objective function after dividing by σ2 is asymptotically χ2 with degrees of freedom equal to the number of overidentifying restrictions. In the present case, there are no overidentifying restrictions, so we have a χ2 (0) rv, which has mean 0 and variance 0, e.g., it’s simply 0. This means we’re not able to test the identifying restrictions in the case of exact identification.

11.8. System methods of estimation 2SLS is a single equation method of estimation, as noted above. The advantage of a single equation method is that it’s unaffected by the other equations of the system,


231

so they don’t need to be specified (except for defining what are the exogs, so 2SLS can use the complete set of instruments). The disadvantage of 2SLS is that it’s inefficient, in general. • Recall that overidentification improves efficiency of estimation, since an overidentified equation can use more instruments than are necessary for consistent estimation. • Secondly, the assumption is that

Y Γ = XB + E E (X 0 E) = 0(K×G)

vec(E) ∼ N(0, Ψ) • Since there is no autocorrelation of the Et ’s, and since the columns of E are individually homoscedastic, then   σ11 In σ12 In · · · σ1G In  ..  σ22 In .  Ψ =  . . ..  . .   · σGG In = Σ ⊗ In

        

This means that the structural equations are heteroscedastic and correlated with one another • In general, ignoring this will lead to inefficient estimation, following the section on GLS. When equations are correlated with one another estimation should account for the correlation in order to obtain efficiency.


232

• Also, since the equations are correlated, information about one equation is implicitly information about all equations. Therefore, overidentification restrictions in any equation improve efficiency for all equations, even the just identified equations. • Single equation methods can’t use these types of information, and are therefore inefficient (in general).

11.8.1. 3SLS. Note: It is easier and more practical to treat the 3SLS estimator as a generalized method of moments estimator (see Chapter 15). I no longer teach the following section, but it is retained for its possible historical interest. Another alternative is to use FIML (Subsection 11.8.2), if you are willing to make distributional assumptions on the errors. This is computationally feasible with modern computers. Following our above notation, each structural equation can be written as yi = Yi γ1 + Xi β1 + εi = Zi δi + ε i

or

Grouping the G equations together we get     y 1   Z1 0 · · · 0    ..  y2   0 Z 2 .     . = . ..  ..   .. . 0       0 · · · 0 ZG yG y = Zδ + ε

        





δ1      δ2    +  ..   .      δG



ε1   ε2   ..  .    εG


233

where we already have that E (εε0 ) = Ψ

= Σ ⊗ In The 3SLS estimator is just 2SLS combined with a GLS correction that takes advantage of the structure of Ψ. Define Zˆ as  0 −1 0  X (X X ) X Z1   0  ˆ Z =  .  ..   0  ˆ 0  Y1 X1   0 Yˆ2  =   ..  .  0 ···

0 X (X 0X )−1 X 0 Z

2

0

···

X2

··· 0 .. . .. . 0

··· 0 .. . .. . 0 0

YˆG XG

X (X 0X )−1 X 0 ZG 

        

       

These instruments are simply the unrestricted rf predicitions of the endogs, com-

bined with the exogs. The distinction is that if the model is overidentified, then Π = BΓ−1 may be subject to some zero restrictions, depending on the restrictions on Γ and B, ˆ does not impose these restrictions. Also, note that Π ˆ is calculated using OLS and Π equation by equation. More on this later. The 2SLS estimator would be δˆ = (Zˆ 0 Z)−1 Zˆ 0 y


234

as can be verified by simple multiplication, and noting that the inverse of a blockdiagonal matrix is just the matrix with the inverses of the blocks on the main diagonal. This IV estimator still ignores the covariance information. The natural extension is to add the GLS transformation, putting the inverse of the error covariance into the formula, which gives the 3SLS estimator δˆ 3SLS = =

Zˆ 0 (Σ ⊗ In )−1 Z

−1

Zˆ 0 (Σ ⊗ In )−1 y

−1 0 −1 Zˆ 0 Σ−1 ⊗ In Z Zˆ Σ ⊗ In y

This estimator requires knowledge of Σ. The solution is to define a feasible estimator using a consistent estimator of Σ. The obvious solution is to use an estimator based on the 2SLS residuals: εˆ i = yi − Zi δˆ i,2SLS (IMPORTANT NOTE: this is calculated using Zi , not Zˆ i ). Then the element i, j of Σ is estimated by σˆ i j =

εˆ 0i εˆ j n

Substitute Σˆ into the formula above to get the feasible 3SLS estimator. Analogously to what we did in the case of 2SLS, the asymptotic distribution of the 3SLS estimator can be shown to be   !−1    −1 0 ˆ ˆ √ ˆ Z (Σ ⊗ In ) Z a  n δ3SLS − δ ∼ N 0, lim E n→∞   n

A formula for estimating the variance of the 3SLS estimator in finite samples (can-

celling out the powers of n) is −1 Vˆ δˆ 3SLS = Zˆ 0 Σˆ −1 ⊗ In Zˆ


235

• This is analogous to the 2SLS formula in equation (??), combined with the GLS correction. • In the case that all equations are just identified, 3SLS is numerically equivalent to 2SLS. Proving this is easiest if we use a GMM interpretation of 2SLS and 3SLS. GMM is presented in the next econometrics course. For now, take it on faith.

ˆ calculated equation The 3SLS estimator is based upon the rf parameter estimator Π, by equation using OLS: ˆ = (X 0 X )−1 X 0Y Π which is simply ˆ = (X 0X )−1 X 0 Π

h

y1 y2 · · · yG

i

that is, OLS equation by equation using all the exogs in the estimation of each column of Π. It may seem odd that we use OLS on the reduced form, since the rf equations are correlated: Yt0 = Xt0 BΓ−1 + Et0 Γ−1 = Xt0 Π +Vt0 and

0 0 Vt = Γ−1 Et ∼ N 0, Γ−1 ΣΓ−1 , ∀t

Let this var-cov matrix be indicated by

0 Ξ = Γ−1 ΣΓ−1


OLS equation by equation to get the rf is equivalent to      y 1   X 0 · · · 0   π1     ..   y2   0 X   π2 .      . = .  . ..  ..   ..   .. . 0         yG 0 ··· 0 X πG





        +      

236



v1   v2   ..  .    vG

where yi is the n × 1 vector of observations of the ith endog, X is the entire n × K matrix

of exogs, πi is the ith column of Π, and vi is the ith column of V. Use the notation y = Xπ + v

to indicate the pooled model. Following this notation, the error covariance matrix is V (v) = Ξ ⊗ In • This is a special case of a type of model known as a set of seemingly unrelated equations (SUR) since the parameter vector of each equation is different. The equations are contemporanously correlated, however. The general case would have a different Xi for each equation. • Note that each equation of the system individually satisfies the classical assumptions. • However, pooled estimation using the GLS correction is more efficient, since equation-by-equation estimation is equivalent to pooled estimation, since X is block diagonal, but ignoring the covariance information. • The model is estimated by GLS, where Ξ is estimated using the OLS residuals from equation-by-equation estimation, which are consistent.


237

• In the special case that all the Xi are the same, which is true in the present case of estimation of the rf parameters, SUR ≡OLS. To show this note that in this case X = In ⊗ X . Using the rules (1) (A ⊗ B)−1 = (A−1 ⊗ B−1 ) (2) (A ⊗ B)0 = (A0 ⊗ B0 ) and (3) (A ⊗ B)(C ⊗ D) = (AC ⊗ BD), we get πˆ SUR = =

=

−1

(In ⊗ X )0 (Ξ ⊗ In )−1 y

−1 −1 Ξ−1 ⊗ X 0 (In ⊗ X ) Ξ ⊗ X0 y

Ξ ⊗ (X 0 X )−1

Ξ−1 ⊗ X 0 y

IG ⊗ (X 0 X )−1 X 0 y    πˆ 1     πˆ 2    =  .   ..      πˆ G

=

(In ⊗ X )0 (Ξ ⊗ In )−1 (In ⊗ X )

• So the unrestricted rf coefficients can be estimated efficiently (assuming normality) by OLS, even if the equations are correlated. • We have ignored any potential zeros in the matrix Π, which if they exist could potentially increase the efficiency of estimation of the rf. • Another example where SUR≡OLS is in estimation of vector autoregressions. See two sections ahead. 11.8.2. FIML. Full information maximum likelihood is an alternative estimation method. FIML will be asymptotically efficient, since ML estimators based on a given information set are asymptotically efficient w.r.t. all other estimators that use the same information set, and in the case of the full-information ML estimator we use the entire


238

information set. The 2SLS and 3SLS estimators don’t require distributional assumptions, while FIML of course does. Our model is, recall Yt0 Γ = Xt0 B + Et0 Et ∼ N(0, Σ), ∀t E (Et Es0 ) = 0,t 6= s

The joint normality of Et means that the density for Et is the multivariate normal, which is (2π)

−g/2

det Σ

−1 −1/2

1 exp − Et0 Σ−1 Et 2

The transformation from Et to Yt requires the Jacobian | det

dEt | = | det Γ| dYt0

so the density for Yt is (2π)

−G/2

| det Γ| det Σ

−1 −1/2

0 1 exp − Yt0 Γ − Xt0 B Σ−1 Yt0 Γ − Xt0 B 2

Given the assumption of independence over time, the joint log-likelihood function is ln L(B, Γ, Σ) = −

0 n 1 n nG ln(2π)+n ln(| det Γ|)− ln det Σ−1 − ∑ Yt0 Γ − Xt0 B Σ−1 Yt0 Γ − Xt0 B 2 2 2 t=1

• This is a nonlinear in the parameters objective function. Maximixation of this can be done using iterative numeric methods. We’ll see how to do this in the next section. • It turns out that the asymptotic distribution of 3SLS and FIML are the same, assuming normality of the errors.

11.9. EXAMPLE: 2SLS AND KLEIN’S MODEL 1

239

• One can calculate the FIML estimator by iterating the 3SLS estimator, thus avoiding the use of a nonlinear optimizer. The steps are (1) Calculate Γˆ 3SLS and Bˆ 3SLS as normal. ˆ = Bˆ 3SLS Γˆ −1 . This is new, we didn’t estimate Π in this way (2) Calculate Π 3SLS before. This estimator may have some zeros in it. When Greene says iterated 3SLS doesn’t lead to FIML, he means this for a procedure that ˆ but only updates Σˆ and Bˆ and Γ. ˆ If you update Π ˆ you doesn’t update Π, do converge to FIML. ˆ and calculate Σˆ using Γˆ and Bˆ to get (3) Calculate the instruments Yˆ = X Π the estimated errors, applying the usual estimator. (4) Apply 3SLS using these new instruments and the estimate of Σ. (5) Repeat steps 2-4 until there is no change in the parameters. • FIML is fully efficient, since it’s an ML estimator that uses all information. This implies that 3SLS is fully efficient when the errors are normally distributed. Also, if each equation is just identified and the errors are normal, then 2SLS will be fully efficient, since in this case 2SLS≡3SLS. • When the errors aren’t normally distributed, the likelihood function is of course different than what’s written above. 11.9. Example: 2SLS and Klein’s Model 1 The Octave program Simeq/Klein.m performs 2SLS estimation for the 3 equations of Klein’s model 1, assuming nonautocorrelated errors, so that lagged endogenous variables can be used as instruments. The results are: CONSUMPTION EQUATION

*******************************************************


240

2SLS estimation results Observations 21 R-squared 0.976711 Sigma-squared 1.044059

estimate

st.err.

t-stat.

p-value

Constant

16.555

1.321

12.534

0.000

Profits

0.017

0.118

0.147

0.885

Lagged Profits

0.216

0.107

2.016

0.060

Wages

0.810

0.040

20.129

0.000

******************************************************* INVESTMENT EQUATION

******************************************************* 2SLS estimation results Observations 21 R-squared 0.884884 Sigma-squared 1.383184

estimate

st.err.

t-stat.

p-value

Constant

20.278

7.543

2.688

0.016

Profits

0.150

0.173

0.867

0.398

Lagged Profits

0.616

0.163

3.784

0.001

Lagged Capital

-0.158

0.036

-4.368

0.000


241

******************************************************* WAGES EQUATION

******************************************************* 2SLS estimation results Observations 21 R-squared 0.987414 Sigma-squared 0.476427

estimate

st.err.

t-stat.

p-value

Constant

1.500

1.148

1.307

0.209

Output

0.439

0.036

12.316

0.000

Lagged Output

0.147

0.039

3.777

0.002

Trend

0.130

0.029

4.475

0.000

*******************************************************

The above results are not valid (specifically, they are inconsistent) if the errors are autocorrelated, since lagged endogenous variables will not be valid instruments in that case. You might consider eliminating the lagged endogenous variables as instruments, and re-estimating by 2SLS, to obtain consistent parameter estimates in this more complex case. Standard errors will still be estimated inconsistently, unless use a Newey-West type covariance estimator. Food for thought...

CHAPTER 12

Introduction to the second half We’ll begin with study of extremum estimators in general. Let Zn be the available data, based on a sample of size n.

D EFINITION 12.0.1. [Extremum estimator] An extremum estimator θˆ is the optimizing element of an objective function sn (Zn , θ) over a set Θ.

We’ll usually write the objective function suppressing the dependence on Z n . Example: Least squares, linear model Let the d.g.p. be yt = xt0 θ0 + εt , t = 1, 2, ..., n, θ0 ∈ Θ. Stacking observations verti 0 0 cally, yn = Xn θ + εn , where Xn = x1 x2 · · · xn . The least squares estimator

is defined as

θˆ ≡ arg min sn (θ) = (1/n) [yn − Xn θ]0 [yn − Xn θ] Θ

We readily find that θˆ = (X0 X)−1 X0 y. Example: Maximum likelihood Suppose that the continuous random variable yt ∼ IIN(θ0 , 1). The maximum likelihood estimator is defined as (yt − θ)2 θˆ ≡ arg max Ln (θ) = ∏ (2π)−1/2 exp − 2 Θ t=1 n

242

!

12. INTRODUCTION TO THE SECOND HALF

243

Because the logarithmic function is strictly increasing on (0, ∞), maximization of the average logarithm of the likelihood function is achieved at the same θˆ as for the likelihood function: n (yt − θ)2 θˆ ≡ arg max sn (θ) = (1/n) ln Ln (θ) = −1/2 ln 2π − (1/n) ∑ Θ 2 t=1

Solution of the f.o.c. leads to the familiar result that θˆ = y¯ . • MLE estimators are asymptotically efficient (Cramér-Rao lower bound, Theorem3), supposing the strong distributional assumptions upon which they are based are true. • One can investigate the properties of an “ML” estimator supposing that the distributional assumptions are incorrect. This gives a quasi-ML estimator, which we’ll study later. • The strong distributional assumptions of MLE may be questionable in many cases. It is possible to estimate using weaker distributional assumptions based only on some of the moments of a random variable(s). Example: Method of moments Suppose we draw a random sample of yt from the χ2 (θ0 ) distribution. Here, θ0 is the parameter of interest. The first moment (expectation), µ1 , of a random variable will in general be a function of the parameters of the distribution, i.e., µ 1 (θ0 ) . • µ1 = µ1 (θ0 ) is a moment-parameter equation.

• In this example, the relationship is the identity function µ1 (θ0 ) = θ0 , though in general the relationship may be more complicated. The sample first moment is µb1 =

n

∑ yt /n.

t=1


244

• Define m1 (θ) = µ1 (θ) − µb1

• The method of moments principle is to choose the estimator of the parameter to set the estimate of the population moment equal to the sample moment, ˆ ≡ 0. Then the moment-parameter equation is inverted to solve for i.e., m1 (θ) the parameter estimate. In this case,

n

ˆ = θˆ − ∑ yt /n = 0. m1 (θ) t=1

p

n Since ∑t=1 yt /n → θ0 by the LLN, the estimator is consistent.

More on the method of moments Continuing with the above example, the variance of a χ2 (θ0 ) r.v. is V (yt ) = E yt − θ0 • Define m2 (θ) = 2θ −

2

= 2θ0 .

n (yt − y) ¯ ∑t=1 n

2

• The MM estimator would set 2

n

¯ ˆ = 2θˆ − ∑t=1 (yt − y) m2 (θ) ≡ 0. n Again, by the LLN, the sample variance is consistent for the true variance, that is,

So,

2

n (yt − y) ¯ p ∑t=1 → 2θ0 . n 2

¯ ∑n (yt − y) , θˆ = t=1 2n

which is obtained by inverting the moment-parameter equation, is consistent.


245

Example: Generalized method of moments (GMM) The previous two examples give two estimators of θ0 which are both consistent. With a given sample, the estimators will be different in general. • With two moment-parameter equations and only one parameter, we have overidentification, which means that we have more information than is strictly necessary for consistent estimation of the parameter. • The GMM combines information from the two moment-parameter equations to form a new estimator which will be more efficient, in general (proof of this below). From the first example, define m1t (θ) = θ − yt . We already have that m1 (θ) is the sample average of m1t (θ), i.e., n

m1 (θ) = 1/n ∑ m1t (θ) t=1 n

= θ − ∑ yt /n. t=1

Clearly, when evaluated at the true parameter value θ0 , both E m1t (θ0 ) = 0 and E m1 (θ0 ) = 0. From the second example we define additional moment conditions m2t (θ) = 2θ − (yt − y) ¯2 and

2

m2 (θ) = 2θ −

n (yt − y) ¯ ∑t=1 . n a.s.

Again, it is clear from the LLN that m2 (θ0 ) → 0. The MM estimator would chose θˆ to

ˆ = 0 or m2 (θ) ˆ = 0. In general, no single value of θ will solve the two set either m1 (θ) equations simultaneously.


246

• The GMM estimator is based on defining a measure of distance d(m(θ)), where m(θ) = (m1 (θ), m2 (θ))0 , and choosing

θˆ = arg min sn (θ) = d (m(θ)) . Θ

An example would be to choose d(m) = m0 Am, where A is a positive definite matrix. While it’s clear that the MM gives consistent estimates if there is a one-to-one relationship between parameters and moments, it’s not immediately obvious that the GMM estimator is consistent. (We’ll see later that it is.) These examples show that these widely used estimators may all be interpreted as the solution of an optimization problem. For this reason, the study of extremum estimators is useful for its generality. We will see that the general results extend smoothly to the more specialized results available for specific estimators. After studying extremum estimators in general, we will study the GMM estimator, then QML and NLS. The reason we study GMM first is that LS, IV, NLS, MLE, QML and other well-known parametric estimators may all be interpreted as special cases of the GMM estimator, so the general results on GMM can simplify and unify the treatment of these other estimators. Nevertheless, there are some special results on QML and NLS, and both are important in empirical research, which makes focus on them useful.

One of the focal points of the course will be nonlinear models. This is not to suggest that linear models aren’t useful. Linear models are more general than they might first appear, since one can employ nonlinear transformations of the variables:

ϕ0 (yt ) = For example,

h

i

0 ϕ1 (xt ) ϕ2 (xt ) · · · ϕ p (xt ) θ + εt

ln yt = α + βx1t + γx21t + δx1t x2t + εt


247

fits this form.

• The important point is that the model is linear in the parameters but not necessarily linear in the variables.

In spite of this generality, situations often arise which simply can not be convincingly represented by linear in the parameters models. Also, theory that applies to nonlinear models also applies to linear models, so one may as well start off with the general case. Example: Expenditure shares Roy’s Identity states that the quantity demanded of the ith of G goods is xi =

−∂v(p, y)/∂pi . ∂v(p, y)/∂y

An expenditure share is si ≡ pi xi /y, so necessarily si ∈ [0, 1], and ∑G i=1 si = 1. No linear in the parameters model for xi or si with a parameter space that is defined independent of the data can guarantee that either of these conditions holds. These constraints will often be violated by estimated linear models, which calls into question their appropriateness in cases of this sort. Example: Binary limited dependent variable The referendum contingent valuation (CV) method of infering the social value of a project provides a simple example. This example is a special case of more general discrete choice (or binary response) models. Individuals are asked if they would pay an amount A for provision of a project. Indirect utility in the base case (no project) is v0 (m, z) + ε0 , where m is income and z is a vector of other variables such as prices, personal characteristics, etc. After provision, utility is v1 (m, z) + ε1 . The random terms εi , i = 1, 2, reflect variations of preferences in the population. With this, an individual


248

agrees1 to pay A if ε|0 {z − ε}1 ε

0 ∂a

for a positive but small. That is, if we go in direction d, we will improve on the objective function, at least if we don’t go too far in that direction. • As long as the gradient at θ is not zero there exist increasing directions, and

they can all be represented as Qk g(θk ) where Qk is a symmetric pd matrix and

13.2. DERIVATIVE-BASED METHODS

254

g (θ) = Dθ s(θ) is the gradient at θ. To see this, take a T.S. expansion around a0 = 0 s(θ + ad) = s(θ + 0d) + (a − 0) g(θ + 0d)0 d + o(1) = s(θ) + ag(θ)0d + o(1) For small enough a the o(1) term can be ignored. If d is to be an increasing direction, we need g(θ)0d > 0. Defining d = Qg(θ), where Q is positive definite, we guarantee that g(θ)0d = g(θ)0 Qg(θ) > 0 unless g(θ) = 0. Every increasing direction can be represented in this way (p.d. matrices are those such that the angle between g and Qg(θ) is less that 90 degrees). See Figure 13.2.1. • With this, the iteration rule becomes θ(k+1) = θ(k) + ak Qk g(θk ) and we keep going until the gradient becomes zero, so that there is no increasing direction. The problem is how to choose a and Q. • Conditional on Q, choosing a is fairly straightforward. A simple line search is an attractive possibility, since a is a scalar. • The remaining problem is how to choose Q. • Note also that this gives no guarantees to find a global maximum. 13.2.2. Steepest descent. Steepest descent (ascent if we’re maximizing) just sets Q to and identity matrix, since the gradient provides the direction of maximum rate of change of the objective function.


255

F IGURE 13.2.1. Increasing directions of search

• Advantages: fast - doesn’t require anything more than first derivatives. • Disadvantages: This doesn’t always work too well however (draw picture of banana function).

13.2.3. Newton-Raphson. The Newton-Raphson method uses information about the slope and curvature of the objective function to determine which direction and how far to move from an initial point. Supposing we’re trying to maximize s n (θ). Take a second order Taylor’s series approximation of sn (θ) about θk (an initial guess). 0 sn (θ) ≈ sn (θk ) + g(θk )0 θ − θk + 1/2 θ − θk H(θk ) θ − θk


256

To attempt to maximize sn (θ), we can maximize the portion of the right-hand side that depends on θ, i.e., we can maximize 0 s(θ) ˜ = g(θk )0 θ + 1/2 θ − θk H(θk ) θ − θk with respect to θ. This is a much easier problem, since it is a quadratic function in θ, so it has linear first order conditions. These are k

k

Dθ s(θ) ˜ = g(θ ) + H(θ ) θ − θ So the solution for the next round estimate is

k

θk+1 = θk − H(θk )−1 g(θk ) This is illustrated in Figure 13.2.2. However, it’s good to include a stepsize, since the approximation to s n (θ) may be ˆ so the actual iteration formula is bad far away from the maximizer θ, θk+1 = θk − ak H(θk )−1 g(θk ) • A potential problem is that the Hessian may not be negative definite when

we’re far from the maximizing point. So −H(θk )−1 may not be positive definite, and −H(θk )−1 g(θk ) may not define an increasing direction of search.

This can happen when the objective function has flat regions, in which case the Hessian matrix is very ill-conditioned (e.g., is nearly singular), or when we’re in the vicinity of a local minimum, H(θk ) is positive definite, and our direction is a decreasing direction of search. Matrix inverses by computers are subject to large errors when the matrix is ill-conditioned. Also, we


257

F IGURE 13.2.2. Newton-Raphson method

certainly don’t want to go in the direction of a minimum when we’re maximizing. To solve this problem, Quasi-Newton methods simply add a positive definite component to H(θ) to ensure that the resulting matrix is positive definite, e.g., Q = −H(θ) + bI, where b is chosen large enough so that Q is well-conditioned and positive definite. This has the benefit that improvement in the objective function is guaranteed. • Another variation of quasi-Newton methods is to approximate the Hessian by using successive gradient evaluations. This avoids actual calculation of the Hessian, which is an order of magnitude (in the dimension of the parameter vector) more costly than calculation of the gradient. They can be done to


258

ensure that the approximation is p.d. DFP and BFGS are two well-known examples. Stopping criteria The last thing we need is to decide when to stop. A digital computer is subject to limited machine precision and round-off errors. For these reasons, it is unreasonable to hope that a program can exactly find the point that maximizes a function. We need to define acceptable tolerances. Some stopping criteria are: • Negligable change in parameters: |θkj − θk−1 j | < ε1 , ∀ j • Negligable relative change: |

θkj − θk−1 j θk−1 j

| < ε2 , ∀ j

• Negligable change of function: |s(θk ) − s(θk−1 )| < ε3 • Gradient negligibly different from zero: |g j (θk )| < ε4 , ∀ j • Or, even better, check all of these. • Also, if we’re maximizing, it’s good to check that the last round (real, not approximate) Hessian is negative definite. Starting values The Newton-Raphson and related algorithms work well if the objective function is concave (when maximizing), but not so well if there are convex regions and local


259

minima or multiple local maxima. The algorithm may converge to a local minimum or to a local maximum that is not optimal. The algorithm may also have difficulties converging at all. • The usual way to “ensure” that a global maximum has been found is to use many different starting values, and choose the solution that returns the highest objective function value. THIS IS IMPORTANT in practice. More on this later. Calculating derivatives The Newton-Raphson algorithm requires first and second derivatives. It is often difficult to calculate derivatives (especially the Hessian) analytically if the function sn (·) is complicated. Possible solutions are to calculate derivatives numerically, or to use programs such as MuPAD or Mathematica to calculate analytic derivatives. For example, Figure 13.2.3 shows MuPAD1 calculating a derivative that I didn’t know off the top of my head, and one that I did know. • Numeric derivatives are less accurate than analytic derivatives, and are usually more costly to evaluate. Both factors usually cause optimization programs to be less successful when numeric derivatives are used. • One advantage of numeric derivatives is that you don’t have to worry about having made an error in calculating the analytic derivative. When programming analytic derivatives it’s a good idea to check that they are correct by using numeric derivatives. This is a lesson I learned the hard way when writing my thesis.

1MuPAD is not a freely distributable program, so it’s not on the CD. You can download it from

http://www.mupad.de/download.shtml


260

F IGURE 13.2.3. Using MuPAD to get analytic derivatives

• Numeric second derivatives are much more accurate if the data are scaled so that the elements of the gradient are of the same order of magnitude. Example: if the model is yt = h(αxt + βzt ) + εt , and estimation is by NLS, suppose that Dα sn (·) = 1000 and Dβ sn (·) = 0.001. One could define α∗ = α/1000; xt∗ = 1000xt ;β∗ = 1000β; zt∗ = zt /1000. In this case, the gradients Dα∗ sn (·) and Dβ sn (·) will both be 1. In general, estimation programs always work better if data is scaled in this way, since roundoff errors are less likely to become important. This is important in practice.

13.4. EXAMPLES

261

• There are algorithms (such as BFGS and DFP) that use the sequential gradient evaluations to build up an approximation to the Hessian. The iterations are faster for this reason since the actual Hessian isn’t calculated, but more iterations usually are required for convergence. • Switching between algorithms during iterations is sometimes useful. 13.3. Simulated Annealing Simulated annealing is an algorithm which can find an optimum in the presence of nonconcavities, discontinuities and multiple local minima/maxima. Basically, the algorithm randomly selects evaluation points, accepts all points that yield an increase in the objective function, but also accepts some points that decrease the objective function. This allows the algorithm to escape from local minima. As more and more points are tried, periodically the algorithm focuses on the best point so far, and reduces the range over which random points are generated. Also, the probability that a negative move is accepted reduces. The algorithm relies on many evaluations, as in the search method, but focuses in on promising areas, which reduces function evaluations with respect to the search method. It does not require derivatives to be evaluated. I have a program to do this if you’re interested.

13.4. Examples This section gives a few examples of how some nonlinear models may be estimated using maximum likelihood. 13.4.1. Discrete Choice: The logit model. In this section we will consider maximum likelihood estimation of the logit model for binary 0/1 dependent variables. We will use the BFGS algotithm to find the MLE.

13.4. EXAMPLES

262

We saw an example of a binary choice model in equation 12.0.1. A more general representation is y∗ = g(x) − ε y = 1(y∗ > 0) Pr(y = 1) = Fε [g(x)] ≡ p(x, θ) The log-likelihood function is 1 n sn (θ) = ∑ (yi ln p(xi , θ) + (1 − yi ) ln [1 − p(xi , θ)]) n i=1 For the logit model (see the contingent valuation example above), the probability has the specific form

p(x, θ) =

1 1 + exp(−x0θ)

You should download and examine LogitDGP.m , which generates data according to the logit model, logit.m , which calculates the loglikelihood, and EstimateLogit.m , which sets things up and calls the estimation routine, which uses the BFGS algorithm. Here are some estimation results with n = 100, and the true θ = (0, 1) 0.

*********************************************** Trial of MLE estimation of Logit model MLE Estimation Results BFGS convergence: Normal convergence Average Log-L: 0.607063 Observations: 100

13.4. EXAMPLES

constant slope

estimate 0.5400 0.7566

st. err 0.2229 0.2374

t-stat 2.4224 3.1863

263

p-value 0.0154 0.0014

Information Criteria CAIC : 132.6230 BIC : 130.6230 AIC : 125.4127 ***********************************************

The estimation program is calling mle_results(), which in turn calls a number of other routines. These functions are part of the octave-forge repository.

13.4.2. Count Data: The Poisson model. Demand for health care is usually thought of a a derived demand: health care is an input to a home production function that produces health, and health is an argument of the utility function. Grossman (1972), for example, models health as a capital stock that is subject to depreciation (e.g., the effects of ageing). Health care visits restore the stock. Under the home production framework, individuals decide when to make health care visits to maintain their health stock, or to deal with negative shocks to the stock in the form of accidents or illnesses. As such, individual demand will be a function of the parameters of the individuals’ utility functions. The MEPS health data file , meps1996.data, contains 4564 observations on six measures of health care usage. The data is from the 1996 Medical Expenditure Panel Survey (MEPS). You can get more information at http://www.meps.ahrq.gov/. The six measures of use are are office-based visits (OBDV), outpatient visits (OPV), inpatient visits (IPV), emergency room visits (ERV), dental visits (VDV), and number of prescription drugs taken (PRESCR). These form columns 1 - 6 of meps1996.data. The conditioning variables are public insurance (PUBLIC), private insurance (PRIV),

13.4. EXAMPLES

264

sex (SEX), age (AGE), years of education (EDUC), and income (INCOME). These form columns 7 - 12 of the file, in the order given here. PRIV and PUBLIC are 0/1 binary variables, where a 1 indicates that the person has access to public or private insurance coverage. SEX is also 0/1, where 1 indicates that the person is female. This data will be used in examples fairly extensively in what follows. The program ExploreMEPS.m shows how the data may be read in, and gives some descriptive information about variables, which follows: All of the measures of use are count data, which means that they take on the values 0, 1, 2, .... It might be reasonable to try to use this information by specifying the density as a count data density. One of the simplest count data densities is the Poisson density, which is fY (y) =

exp(−λ)λy . y!

The Poisson average log-likelihood function is 1 n sn (θ) = ∑ (−λi + yi ln λi − ln yi !) n i=1 We will parameterize the model as λi = exp(x0i β) xi = [1 PU BLIC PRIV SEX AGE EDUC INC]0. This ensures that the mean is positive, as is required for the Poisson model. Note that for this parameterization βj = so

∂λ/∂β j λ

13.4. EXAMPLES

265

β j x j = ηλx j , the elasticity of the conditional mean of y with respect to the jth conditioning variable. The program EstimatePoisson.m estimates a Poisson model using the full data set. The results of the estimation, using OBDV as the dependent variable are here:

MPITB extensions found

OBDV

****************************************************** Poisson model, MEPS 1996 full data set

MLE Estimation Results BFGS convergence: Normal convergence

Average Log-L: -3.671090 Observations: 4564

estimate

st. err

t-stat

p-value

constant

-0.791

0.149

-5.290

0.000

pub. ins.

0.848

0.076

11.093

0.000

priv. ins.

0.294

0.071

4.137

0.000

sex

0.487

0.055

8.797

0.000

age

0.024

0.002

11.471

0.000

13.4. EXAMPLES

266

edu

0.029

0.010

3.061

0.002

inc

-0.000

0.000

-0.978

0.328

Information Criteria CAIC : 33575.6881

Avg. CAIC:

7.3566

BIC : 33568.6881

Avg. BIC:

7.3551

AIC : 33523.7064

Avg. AIC:

7.3452

******************************************************

13.4.3. Duration data and the Weibull model. In some cases the dependent variable may be the time that passes between the occurence of two events. For example, it may be the duration of a strike, or the time needed to find a job once one is unemployed. Such variables take on values on the positive real line, and are referred to as duration data. A spell is the period of time between the occurence of initial event and the concluding event. For example, the initial event could be the loss of a job, and the final event is the finding of a new job. The spell is the period of unemployment. Let t0 be the time the initial event occurs, and t1 be the time the concluding event occurs. For simplicity, assume that time is measured in years. The random variable D is the duration of the spell, D = t1 − t0 . Define the density function of D, f D (t), with distribution function FD (t) = Pr(D < t). Several questions may be of interest. For example, one might wish to know the expected time one has to wait to find a job given that one has already waited s years. The probability that a spell lasts s years is Pr(D > s) = 1 − Pr(D ≤ s) = 1 − FD (s).

13.4. EXAMPLES

267

The density of D conditional on the spell already having lasted s years is

fD (t|D > s) =

fD (t) . 1 − FD (s)

The expectanced additional time required for the spell to end given that is has already lasted s years is the expectation of D with respect to this density, minus s. E = E (D|D > s) − s =

Z

t

∞

fD (z) dz − s z 1 − FD (s)

To estimate this function, one needs to specify the density f D (t) as a parametric density, then estimate by maximum likelihood. There are a number of possibilities including the exponential density, the lognormal, etc. A reasonably flexible model that is a generalization of the exponential density is the Weibull density γ

fD (t|θ) = e−(λt) λγ(λt)γ−1. According to this model, E (D) = λ−γ . The log-likelihood is just the product of the log densities. To illustrate application of this model, 402 observations on the lifespan of mongooses in Serengeti National Park (Tanzania) were used to fit a Weibull model. The ”spell” in this case is the lifetime of an individual mongoose. The parameter estimates ˆ = 0.559 (0.034) and γˆ = 0.867 (0.033) and the log-likelihood and standard errors are λ value is -659.3. Figure 13.4.1 presents fitted life expectancy (expected additional years of life) as a function of age, with 95% confidence bands. The plot is accompanied by a nonparametric Kaplan-Meier estimate of life-expectancy. This nonparametric estimator simply averages all spell lengths greater than age, and then subtracts age. This is consistent by the LLN.

13.4. EXAMPLES

268

F IGURE 13.4.1. Life expectancy of mongooses, Weibull model

In the figure one can see that the model doesn’t fit the data well, in that it predicts life expectancy quite differently than does the nonparametric model. For ages 4-6, the nonparametric estimate is outside the confidence interval that results from the parametric model, which casts doubt upon the parametric model. Mongooses that are between 2-6 years old seem to have a lower life expectancy than is predicted by the Weibull model, whereas young mongooses that survive beyond infancy have a higher life expectancy, up to a bit beyond 2 years. Due to the dramatic change in the death

13.4. EXAMPLES

269

rate as a function of t, one might specify f D (t) as a mixture of two Weibull densities,

fD (t|θ) = δ e

−(λ1t)γ1

λ1 γ1 (λ1t)

γ1 −1

+ (1 − δ) e

−(λ2t)γ2

λ2 γ2 (λ2t)

γ2 −1

.

The parameters γi and λi , i = 1, 2 are the parameters of the two Weibull densities, and δ is the parameter that mixes the two. With the same data, θ can be estimated using the mixed model. The results are a log-likelihood = -623.17. Note that a standard likelihood ratio test cannot be used to chose between the two models, since under the null that δ = 1 (single density), the two parameters λ2 and γ2 are not identified. It is possible to take this into account, but this topic is out of the scope of this course. Nevertheless, the improvement in the likelihood function is considerable. The parameter estimates are Parameter Estimate St. Error λ1

0.233

0.016

γ1

1.722

0.166

λ2

1.731

0.101

γ2

1.522

0.096

δ

0.428

0.035

Note that the mixture parameter is highly significant. This model leads to the fit in Figure 13.4.2. Note that the parametric and nonparametric fits are quite close to one another, up to around 6 years. The disagreement after this point is not too important, since less than 5% of mongooses live more than 6 years, which implies that the KaplanMeier nonparametric estimate has a high variance (since it’s an average of a small number of observations). Mixture models are often an effective way to model complex responses, though they can suffer from overparameterization. Alternatives will be discussed later.

13.5. NUMERIC OPTIMIZATION: PITFALLS

270

F IGURE 13.4.2. Life expectancy of mongooses, mixed Weibull model

13.5. Numeric optimization: pitfalls In this section we’ll examine two common problems that can be encountered when doing numeric optimization of nonlinear models, and some solutions. 13.5.1. Poor scaling of the data. When the data is scaled so that the magnitudes of the first and second derivatives are of different orders, problems can easily result. If we uncomment the appropriate line in EstimatePoisson.m, the data will not be scaled, and the estimation program will have difficulty converging (it seems to take an infinite amount of time). With unscaled data, the elements of the score vector have very different magnitudes at the initial value of θ (all zeros). To see this run CheckScore.m. With


271

unscaled data, one element of the gradient is very large, and the maximum and minimum elements are 5 orders of magnitude apart. This causes convergence problems due to serious numerical inaccuracy when doing inversions to calculate the BFGS direction of search. With scaled data, none of the elements of the gradient are very large, and the maximum difference in orders of magnitude is 3. Convergence is quick.

13.5.2. Multiple optima. Multiple optima (one global, others local) can complicate life, since we have limited means of determining if there is a higher maximum the the one we’re at. Think of climbing a mountain in an unknown range, in a very foggy place (Figure 13.5.1). You can go up until there’s nowhere else to go up, but since you’re in the fog you don’t know if the true summit is across the gap that’s at your feet. Do you claim victory and go home, or do you trudge down the gap and explore the other side? The best way to avoid stopping at a local maximum is to use many starting values, for example on a grid, or randomly generated. Or perhaps one might have priors about possible values for the parameters (e.g., from previous studies of similar data). Let’s try to find the true minimizer of minus 1 times the foggy mountain function (since the algoritms are set up to minimize). From the picture, you can see it’s close to (0, 0), but let’s pretend there is fog, and that we don’t know that. The program FoggyMountain.m shows that poor start values can lead to problems. It uses SA, which finds the true global minimum, and it shows that BFGS using a battery of random start values can also find the global minimum help. The output of one run is here: MPITB extensions found

====================================================== BFGSMIN final results


F IGURE 13.5.1. A foggy mountain

Used numeric gradient

-----------------------------------------------------STRONG CONVERGENCE Function conv 1

Param conv 1

Gradient conv 1

-----------------------------------------------------Objective function value -0.0130329 Stepsize 0.102833 43 iterations

272


------------------------------------------------------

param

gradient

change

15.9999

-0.0000

0.0000

-28.8119

0.0000

0.0000

The result with poor start values ans =

16.000

-28.812

================================================ SAMIN final results NORMAL CONVERGENCE

Func. tol. 1.000000e-10 Param. tol. 1.000000e-03 Obj. fn. value -0.100023

parameter

search width

0.037419

0.000018

-0.000000

0.000051

================================================ Now try a battery of random start values and a short BFGS on each, then iterate to convergence The result using 20 randoms start values ans =

273


3.7417e-02

274

2.7628e-07

The true maximizer is near (0.037,0)

In that run, the single BFGS run with bad start values converged to a point far from the true minimizer, which simulated annealing and BFGS using a battery of random start values both found the true maximizaer. battery of random start values managed to find the global max. The moral of the story is be cautious and don’t publish your results too quickly.

EXERCISES

275

Exercises (1) In octave, type ”help bfgsmin_example”, to find out the location of the file. Edit the file to examine it and learn how to call bfgsmin. Run it, and examine the output. (2) In octave, type ”help samin_example”, to find out the location of the file. Edit the file to examine it and learn how to call samin. Run it, and examine the output. (3) Using logit.m and EstimateLogit.m as templates, write a function to calculate the probit loglikelihood, and a script to estimate a probit model. Run it using data that actually follows a logit model (you can generate it in the same way that is done in the logit example). (4) Study mle_results.m to see what it does. Examine the functions that mle_results.m calls, and in turn the functions that those functions call. Write a complete description of how the whole chain works. (5) Look at the Poisson estimation results for the OBDV measure of health care use and give an economic interpretation. Estimate Poisson models for the other 5 measures of health care usage.

CHAPTER 14

Asymptotic properties of extremum estimators Readings: Gourieroux and Monfort (1995), Vol. 2, Ch. 24∗ ; Amemiya, Ch. 4 section 4.1∗ ; Davidson and MacKinnon, pp. 591-96; Gallant, Ch. 3; Newey and McFadden (1994), “Large Sample Estimation and Hypothesis Testing,” in Handbook of Econometrics, Vol. 4, Ch. 36.

14.1. Extremum estimators In Definition 12.0.1 we defined an extremum estimator θˆ as the optimizing element of an objective function sn (θ) over a set Θ. Let the objective function sn (Zn , θ) depend h i0 upon a n × p random matrix Zn = z1 z2 · · · zn where the zt are p-vectors and p is finite.

E XAMPLE 18. Given the model yi = x0i θ + εi , with n observations, define zi = (yi , x0i )0 . The OLS estimator minimizes n

sn (Zn , θ) = 1/n ∑ yi − x0i θ i=1

2

= 1/n k Y − X θ k2 where Y and X are defined similarly to Z. 276

14.2. CONSISTENCY

277

14.2. Consistency The following theorem is patterned on a proof in Gallant (1987) (the article, ref. later), which we’ll see in its original form later in the course. It is interesting to compare the following proof with Amemiya’s Theorem 4.1.1, which is done in terms of convergence in probability. T HEOREM 19. [Consistency of e.e.] Suppose that θˆ n is obtained by maximizing sn (θ) over Θ. Assume (1) Compactness: The parameter space Θ is an open bounded subset of Euclidean space ℜK . So the closure of Θ, Θ, is compact. (2) Uniform Convergence: There is a nonstochastic function s∞ (θ) that is continuous in θ on Θ such that lim sup |sn (θ) − s∞ (θ)| = 0, a.s.

n→∞

θ∈Θ

(3) Identification: s∞ (·) has a unique global maximum at θ0 ∈ Θ, i.e., s∞ (θ0 ) > s∞ (θ), ∀θ 6= θ0 , θ ∈ Θ

a.s.

Then θˆ n → θ0 . Proof: Select a ω ∈ Ω and hold it fixed. Then {sn (ω, θ)} is a fixed sequence of

functions. Suppose that ω is such that sn (θ) converges uniformly to s∞ (θ). This happens with probability one by assumption (b). The sequence { θˆ n } lies in the compact set Θ, by assumption (1) and the fact that maximixation is over Θ. Since every se-

quence from a compact set has at least one limit point (Davidson, Thm. 2.12), say that θˆ is a limit point of {θˆ n }. There is a subsequence {θˆ nm } ({nm } is simply a sequence of

14.2. CONSISTENCY

278

ˆ By uniform convergence and continuity increasing integers) with limm→∞ θˆ nm = θ. ˆ lim snm (θˆ nm ) = s∞ (θ).

m→∞

To see this, first of all, select an element θˆ t from the sequence θˆ nm . Then uniform convergence implies

lim snm (θˆ t ) = s∞ (θˆ t ).

m→∞

Continuity of s∞ (·) implies that ˆ lim s∞ (θˆ t ) = s∞ (θ)

t→∞

ˆ So the above claim is true. since the limit as t → ∞ of θˆ t is θ. Next, by maximization

snm (θˆ nm ) ≥ snm (θ0 ) which holds in the limit, so lim snm (θˆ nm ) ≥ lim snm (θ0 ).

m→∞

m→∞

However, ˆ lim snm (θˆ nm ) = s∞ (θ),

m→∞

as seen above, and lim snm (θ0 ) = s∞ (θ0 )

m→∞

by uniform convergence, so ˆ ≥ s∞ (θ0 ). s∞ (θ) But by assumption (3), there is a unique global maximum of s∞ (θ) at θ0 , so we must ˆ = s∞ (θ0 ), and θˆ = θ0 . Finally, all of the above limits hold almost surely, have s∞ (θ)

14.2. CONSISTENCY

279

since so far we have held ω fixed, but now we need to consider all ω ∈ Ω. Therefore {θˆ n } has only one limit point, θ0 , except on a set C ⊂ Ω with P(C) = 0. Discussion of the proof: • This proof relies on the identification assumption of a unique global maximum at θ0 . An equivalent way to state this is

(2) Identification: Any point θ in Θ with s∞ (θ) ≥ s∞ (θ0 ) must be such that k θ − θ0 k= 0, which matches the way we will write the assumption in the section on nonparametric inference. • We assume that θˆ n is in fact a global maximum of sn (θ) . It is not required to be unique for n finite, though the identification assumption requires that the limiting objective function have a unique maximizing argument. The previous section on numeric optimization methods showed that actually finding the global maximum of sn (θ) may be a non-trivial problem. • See Amemiya’s Example 4.1.4 for a case where discontinuity leads to breakdown of consistency. • The assumption that θ0 is in the interior of Θ (part of the identification assumption) has not been used to prove consistency, so we could directly assume that θ0 is simply an element of a compact set Θ. The reason that we assume it’s in the interior here is that this is necessary for subsequent proof of asymptotic normality, and I’d like to maintain a minimal set of simple assumptions, for clarity. Parameters on the boundary of the parameter set cause theoretical difficulties that we will not deal with in this course. Just note that conventional hypothesis testing methods do not apply in this case. • Note that sn (θ) is not required to be continuous, though s∞ (θ) is. • The following figures illustrate why uniform convergence is important. In the second figure, if the function is not converging around the lower of the two

14.2. CONSISTENCY

280

maxima, there is no guarantee that the maximizer will be in the neighborhood of the global maximizer.

With uniform convergence, the maximum of the sample objective function eventually must be in the neighborhood of the maximum of the limiting objective function

With pointwise convergence, the sample objective function may have its maximum far away from that of the limiting objective function

14.2. CONSISTENCY

281

We need a uniform strong law of large numbers in order to verify assumption (2) of Theorem 19. The following theorem is from Davidson, pg. 337. T HEOREM 20. [Uniform Strong LLN] Let {Gn (θ)} be a sequence of stochastic

real-valued functions on a totally-bounded metric space (Θ, ρ). Then a.s.

sup |Gn (θ)| → 0

θ∈Θ

if and only if a.s.

(a) Gn (θ) → 0 for each θ ∈ Θ0 , where Θ0 is a dense subset of Θ and (b) {Gn (θ)} is strongly stochastically equicontinuous.. • The metric space we are interested in now is simply Θ ⊂ ℜK , using the Euclidean norm. • The pointwise almost sure convergence needed for assuption (a) comes from one of the usual SLLN’s. • Stronger assumptions that imply those of the theorem are: – the parameter space is compact (this has already been assumed) – the objective function is continuous and bounded with probability one on the entire parameter space – a standard SLLN can be shown to apply to some point in the parameter space • These are reasonable conditions in many cases, and henceforth when dealing with specific estimators we’ll simply assume that pointwise almost sure convergence can be extended to uniform almost sure convergence in this way. • The more general theorem is useful in the case that the limiting objective function can be continuous in θ even if sn (θ) is discontinuous. This can happen because discontinuities may be smoothed out as we take expectations

14.3. EXAMPLE: CONSISTENCY OF LEAST SQUARES

282

over the data. In the section on simlation-based estimation we will se a case of a discontinuous objective function.

14.3. Example: Consistency of Least Squares We suppose that data is generated by random sampling of (y, w), where yt = α0 + β0 wt +εt . (wt , εt ) has the common distribution function µw µε (w and ε are independent) with support W × E . Suppose that the variances σ2w and σ2ε are finite.

Let θ0 = (α0 , β0 )0 ∈ Θ, for which Θ is compact. Let xt = (1, wt )0 , so we can write

yt = xt0 θ0 + εt . The sample objective function for a sample size n is n

sn (θ) = 1/n ∑ yt − xt0 θ t=1 n

2

= 1/n ∑ xt0 θ0 − θ t=1

n

= 1/n ∑ xt0 θ0 + εt − xt0 θ i=1

2

2

n n + 2/n ∑ xt0 θ0 − θ εt + 1/n ∑ εt2 t=1

t=1

• Considering the last term, by the SLLN, n

1/n ∑

t=1

a.s. εt2 →

Z

W

Z

E

ε2 dµW dµE = σ2ε .

• Considering the second term, since E(ε) = 0 and w and ε are independent, the SLLN implies that it converges to zero. • Finally, for the first term, for a given θ, we assume that a SLLN applies so that n

1/n ∑

(14.3.1)

t=1

= =

α0 − α α0 − α

xt0 2

2

2 a.s. Z → θ −θ 0

+ 2 α0 − α

+ 2 α0 − α

W

x0 θ 0 − θ

β0 − β

Z

W

2

dµW

wdµW + β0 − β 2

2 Z

β0 − β E(w) + β0 − β E w

W

2

w2 dµW


283

Finally, the objective function is clearly continuous, and the parameter space is assumed to be compact, so the convergence is also uniform. Thus, s∞ (θ) = α0 − α

2

+ 2 α0 − α

2 β0 − β E(w) + β0 − β E w2 + σ2ε

A minimizer of this is clearly α = α0 , β = β0 .

E XERCISE 21. Show that in order for the above solution to be unique it is necessary that E(w2 ) 6= 0. Discuss the relationship between this condition and the problem of colinearity of regressors. This example shows that Theorem 19 can be used to prove strong consistency of the OLS estimator. There are easier ways to show this, of course - this is only an example of application of the theorem. 14.4. Asymptotic Normality A consistent estimator is oftentimes not very useful unless we know how fast it is likely to be converging to the true value, and the probability that it is far away from the true value. Establishment of asymptotic normality with a known scaling factor solves these two problems. The following theorem is similar to Amemiya’s Theorem 4.1.3 (pg. 111). T HEOREM 22. [Asymptotic normality of e.e.] In addition to the assumptions of Theorem 19, assume

θ0 .

(a) Jn (θ) ≡ D2θ sn (θ) exists and is continuous in an open, convex neighborhood of a.s.

(b) {Jn (θn )} → J∞ (θ0 ), a finite negative definite matrix, for any sequence {θn }

that converges almost surely to θ0 . √ √ d (c) nDθ sn (θ0 ) → N 0, I∞ (θ0 ) , where I∞ (θ0 ) = limn→∞ Var nDθ sn (θ0 )


Then

√

284

d n θˆ − θ0 → N 0, J∞ (θ0 )−1 I∞ (θ0 )J∞ (θ0 )−1

Proof: By Taylor expansion: Dθ sn (θˆ n ) = Dθ sn (θ0 ) + D2θ sn (θ∗ ) θˆ − θ0 where θ∗ = λθˆ + (1 − λ)θ0 , 0 ≤ λ ≤ 1.

• Note that θˆ will be in the neighborhood where D2θ sn (θ) exists with probability one as n becomes large, by consistency. • Now the l.h.s. of this equation is zero, at least asymptotically, since θˆ n is a maximizer and the f.o.c. must hold exactly since the limiting objective function is strictly concave in a neighborhood of θ0 . a.s.

• Also, since θ∗ is between θˆ n and θ0 , and since θˆ n → θ0 , assumption (b) gives a.s.

D2θ sn (θ∗ ) → J∞ (θ0 ) So

And

0 = Dθ sn (θ0 ) + J∞ (θ0 ) + o p (1) θˆ − θ0 0=

√ √ nDθ sn (θ0 ) + J∞ (θ0 ) + o p (1) n θˆ − θ0

Now J∞ (θ0 ) is a finite negative definite matrix, so the o p (1) term is asymptotically irrelevant next to J∞ (θ0 ), so we can write a

0= √

√ √ nDθ sn (θ0 ) + J∞ (θ0 ) n θˆ − θ0

a √ n θˆ − θ0 = −J∞ (θ0 )−1 nDθ sn (θ0 )


285

Because of assumption (c), and the formula for the variance of a linear combination of r.v.’s, √

d n θˆ − θ0 → N 0, J∞ (θ0 )−1 I∞ (θ0 )J∞ (θ0 )−1

• Assumption (b) is not implied by the Slutsky theorem. The Slutsky theorem a.s.

says that g(xn ) → g(x) if xn → x and g(·) is continuous at x. However, the function g(·) can’t depend on n to use this theorem. In our case J n (θn ) is a function of n. A theorem which applies (Amemiya, Ch. 4) is T HEOREM 23. If gn (θ) converges uniformly almost surely to a nonstochastic funca.s.

ˆ → g∞ (θ0 ) if g∞ (θ0 ) tion g∞ (θ) uniformly on an open neighborhood of θ0 , then gn (θ) a.s. is continuous at θ0 and θˆ → θ0 .

• To apply this to the second derivatives, sufficient conditions would be that the second derivatives be strongly stochastically equicontinuous on a neighborhood of θ0 , and that an ordinary LLN applies to the derivatives when evaluated at θ ∈ N(θ0 ). • Stronger conditions that imply this are as above: continuous and bounded second derivatives in a neighborhood of θ0 .

• Skip this in lecture. A note on the order of these matrices: Supposing that sn (θ) is representable as an average of n terms, which is the case for all estimators we consider, D2θ sn (θ) is also an average of n matrices, the elements of which are not centered (they do not have zero expectation). Supposing a SLLN applies, the almost sure limit of D2θ sn (θ0 ), J∞ (θ0 ) = O(1), as we saw in √ d Example 51. On the other hand, assumption (c): nDθ sn (θ0 ) → N 0, I∞ (θ0 ) means that

√ nDθ sn (θ0 ) = O p ()

14.5. EXAMPLES

286

where we use the result of Example 49. If we were to omit the

√

n, we’d have

1

Dθ sn (θ0 ) = n− 2 O p (1) 1 = O p n− 2 where we use the fact that O p (nr )O p (nq ) = O p (nr+q ). The sequence Dθ sn (θ0 ) √ is centered, so we need to scale by n to avoid convergence to zero.

14.5. Examples 14.5.1. Binary response models. Binary response models arise in a variety of contexts. We’ve already seen a logit model. Another simple example is a probit threshold-crossing model. Assume that y∗ = x 0 β − ε y = 1(y∗ > 0) ε ∼ N(0, 1) Here, y∗ is an unobserved (latent) continuous variable, and y is a binary variable that indicates whether y∗ is negative or positive. Then Pr(y = 1) = Pr(ε < xβ) = Φ(xβ), where Φ(•) =

Z xβ −∞

(2π)−1/2 exp(−

ε2 )dε 2

is the standard normal distribution function. In general, a binary response model will require that the choice probability be parameterized in some form. For a vector of explanatory variables x, the response probability will be parameterized in some manner Pr(y = 1|x) = p(x, θ)

14.5. EXAMPLES

287

If p(x, θ) = Λ(x0 θ), we have a logit model. If p(x, θ) = Φ(x0 θ), where Φ(·) is the standard normal distribution function, then we have a probit model. Regardless of the parameterization, we are dealing with a Bernoulli density, fYi (yi |xi ) = p(xi , θ)yi (1 − p(x, θ))1−yi so as long as the observations are independent, the maximum likelihood (ML) estimaˆ is the maximizer of tor, θ, sn (θ) = (14.5.1)

≡

1 n ∑ (yi ln p(xi, θ) + (1 − yi) ln [1 − p(xi, θ)]) n i=1 1 n ∑ s(yi, xi, θ). n i=1

Following the above theoretical results, θˆ tends in probability to the θ0 that maximizes the uniform almost sure limit of sn (θ). Noting that E yi = p(xi , θ0 ), and following a SLLN for i.i.d. processes, sn (θ) converges almost surely to the expectation of a representative term s(y, x, θ). First one can take the expectation conditional on x to get

Ey|x {y ln p(x, θ) + (1 − y) ln[1 − p(x, θ)]} = p(x, θ0 ) ln p(x, θ)+ 1 − p(x, θ0 ) ln [1 − p(x, θ)] .

Next taking expectation over x we get the limiting objective function (14.5.2)

s∞ (θ) =

Z X

p(x, θ0 ) ln p(x, θ) + 1 − p(x, θ0 ) ln [1 − p(x, θ)] µ(x)dx,

where µ(x) is the (joint - the integral is understood to be multiple, and X is the support of x) density function of the explanatory variables x. This is clearly continuous in θ, as long as p(x, θ) is continuous, and if the parameter space is compact we therefore have uniform almost sure convergence. Note that p(x, θ) is continous for the logit and probit models, for example. The maximizing element of s∞ (θ), θ∗ , solves the first

14.5. EXAMPLES

288

order conditions Z X

p(x, θ0 ) ∂ 1 − p(x, θ0 ) ∂ ∗ ∗ p(x, θ ) − p(x, θ ) µ(x)dx = 0 p(x, θ∗ ) ∂θ 1 − p(x, θ∗ ) ∂θ

This is clearly solved by θ∗ = θ0 . Provided the solution is unique, θˆ is consistent. Question: what’s needed to ensure that the solution is unique? The asymptotic normality theorem tells us that d √ n θˆ − θ0 → N 0, J∞ (θ0 )−1 I∞ (θ0 )J∞ (θ0 )−1 .

√ In the case of i.i.d. observations I∞ (θ0 ) = limn→∞ Var nDθ sn (θ0 ) is simply the expectation of a typical element of the outer product of the gradient. • There’s no need to subtract the mean, since it’s zero, following the f.o.c. in the consistency proof above and the fact that observations are i.i.d. • The terms in n also drop out by the same argument: √ lim Var nDθ sn (θ0 ) =

n→∞

√ 1 lim Var nDθ ∑ s(θ0 ) n→∞ n t

=

1 lim Var √ Dθ ∑ s(θ0 ) n→∞ n t

=

1 lim Var ∑ Dθ s(θ0 ) n→∞ n t

=

lim VarDθ s(θ0 )

n→∞

= VarDθ s(θ0 ) So we get 0

I∞ (θ ) = E

∂ 0 ∂ 0 s(y, x, θ ) 0 s(y, x, θ ) . ∂θ ∂θ

Likewise, J∞ (θ0 ) = E

∂2 s(y, x, θ0 ). ∂θ∂θ0

14.5. EXAMPLES

289

Expectations are jointly over y and x, or equivalently, first over y conditional on x, then over x. From above, a typical element of the objective function is s(y, x, θ0 ) = y ln p(x, θ0 ) + (1 − y) ln 1 − p(x, θ0 ) .

Now suppose that we are dealing with a correctly specified logit model: p(x, θ) = 1 + exp(−x0 θ)

−1

.

We can simplify the above results in this case. We have that

∂ p(x, θ) = ∂θ

1 + exp(−x0 θ)

=

1 + exp(−x0 θ)

−2

−1

exp(−x0 θ)x exp(−x0 θ) x 1 + exp(−x0 θ)

= p(x, θ) (1 − p(x, θ)) x = So

p(x, θ) − p(x, θ)2 x.

∂ s(y, x, θ0 ) = y − p(x, θ0 ) x ∂θ 0 ∂2 0 0 0 2 s(θ ) = − p(x, θ ) − p(x, θ ) xx . ∂θ∂θ0

(14.5.3)

Taking expectations over y then x gives (14.5.4) (14.5.5)

0

I∞ (θ ) =

=

Z

EY y2 − 2p(x, θ0 )p(x, θ0 ) + p(x, θ0 )2 xx0 µ(x)dx

Z

p(x, θ0 ) − p(x, θ0 )2 xx0 µ(x)dx.

14.5. EXAMPLES

290

where we use the fact that EY (y) = EY (y2 ) = p(x, θ0 ). Likewise, (14.5.6)

0

J∞ (θ ) = −

Z

p(x, θ0 ) − p(x, θ0 )2 xx0 µ(x)dx.

Note that we arrive at the expected result: the information matrix equality holds (that is, J∞ (θ0 ) = −I∞ (θ0 )). With this,

simplifies to

d √ n θˆ − θ0 → N 0, J∞ (θ0 )−1 I∞ (θ0 )J∞ (θ0 )−1 d √ n θˆ − θ0 → N 0, −J∞ (θ0 )−1

which can also be expressed as √

d n θˆ − θ0 → N 0, I∞ (θ0 )−1 .

On a final note, the logit and standard normal CDF’s are very similar - the logit distribution is a bit more fat-tailed. While coefficients will vary slightly between the two ˆ will be virtually models, functions of interest such as estimated probabilities p(x, θ) identical for the two models.

14.5.2. Example: Linearization of a nonlinear model. Ref. Gourieroux and Monfort, section 8.3.4. White, Intn’l Econ. Rev. 1980 is an earlier reference. Suppose we have a nonlinear model yi = h(xi , θ0 ) + εi where εi ∼ iid(0, σ2 )

14.5. EXAMPLES

291

The nonlinear least squares estimator solves n

1 θˆ n = arg min ∑ (yi − h(xi , θ))2 n i=1 We’ll study this more later, but for now it is clear that the foc for minimization will require solving a set of nonlinear equations. A common approach to the problem seeks to avoid this difficulty by linearizing the model. A first order Taylor’s series expansion about the point x0 with remainder gives yi = h(x0 , θ0 ) + (xi − x0 )0

∂h(x0 , θ0 ) + νi ∂x

where νi encompasses both εi and the Taylor’s series remainder. Note that νi is no longer a classical error - its mean is not zero. We should expect problems. Define α∗ = h(x0 , θ0 ) − x00 β∗ =

∂h(x0 , θ0 ) ∂x

∂h(x0 , θ0 ) ∂x

Given this, one might try to estimate α∗ and β∗ by applying OLS to yi = α + βxi + νi • Question, will αˆ and βˆ be consistent for α∗ and β∗ ?

• The answer is no, as one can see by interpreting αˆ and βˆ as extremum estimators. Let γ = (α, β0 )0 .

γˆ = arg min sn (γ) =

1 n ∑ (yi − α − βxi)2 n i=1

The objective function converges to its expectation u.a.s.

sn (γ) → s∞ (γ) = EX EY |X (y − α − βx)2

14.5. EXAMPLES

292

and γˆ converges a.s. to the γ0 that minimizes s∞ (γ): γ0 = arg min EX EY |X (y − α − βx)2 Noting that EX EY |X y − α − x0 β

2

= EX EY |X h(x, θ0 ) + ε − α − βx = σ2 + EX h(x, θ0 ) − α − βx

2

2

since cross products involving ε drop out. α0 and β0 correspond to the hyperplane that is closest to the true regression function h(x, θ0 ) according to the mean squared error criterion. This depends on both the shape of h(·) and the density function of the conditioning variables. Inconsistency of the linear approximation, even at the approximation point x

h(x,θ) x

Tangent line

x β

α

x

x

x

x

x

Fitted line

x x

x_0

• It is clear that the tangent line does not minimize MSE, since, for example, if

h(x, θ0 ) is concave, all errors between the tangent line and the true function are negative.

14.5. EXAMPLES

293

• Note that the true underlying parameter θ0 is not estimated consistently, either (it may be of a different dimension than the dimension of the parameter of the approximating model, which is 2 in this example). • Second order and higher-order approximations suffer from exactly the same problem, though to a less severe degree, of course. For this reason, translog, Generalized Leontiev and other “flexible functional forms” based upon secondorder approximations in general suffer from bias and inconsistency. The bias may not be too important for analysis of conditional means, but it can be very important for analyzing first and second derivatives. In production and consumer analysis, first and second derivatives (e.g., elasticities of substitution) are often of interest, so in this case, one should be cautious of unthinking application of models that impose stong restrictions on second derivatives. • This sort of linearization about a long run equilibrium is a common practice in dynamic macroeconomic models. It is justified for the purposes of theoretical analysis of a model given the model’s parameters, but it is not justifiable for the estimation of the parameters of the model using data. The section on simulation-based methods offers a means of obtaining consistent estimators of the parameters of dynamic macro models that are too complex for standard methods of analysis.

14.5. EXAMPLES

294

Chapter Exercises (1) Suppose that xi ∼ uniform(0,1), and yi = 1 − x2i + εi , where εi is iid(0,σ2 ). Suppose we estimate the misspecified model yi = α + βxi + ηi by OLS. Find the numeric values of α0 and β0 that are the probability limits of αˆ and βˆ (2) Verify your results using Octave by generating data that follows the above model, and calculating the OLS estimator. When the sample size is very large the estimator should be very close to the analytical results you obtained in question 1. (3) Use the asymptotic normality theorem to find the asymptotic distribution of the ML estimator of β0 for the model y = xβ0 + ε, where ε ∼ N(0, 1) and is ∂sn (β) ∂2 0 independent of x. This means finding ∂β∂β0 sn (β), J (β ), ∂β , and I (β0 ). The expressions may involve the unspecified density of x.

(4) Assume a d.g.p. follows the logit model: Pr(y = 1|x) = 1 + exp(−β0 x)

−1

.

(a) Assume that x ∼ uniform(-a,a). Find the asymptotic distribution of the ML estimator of β0 (this is a scalar parameter).

(b) Now assume that x ∼ uniform(-2a,2a). Again find the asymptotic distribution of the ML estimator of β0 .

(c) Comment on the results

CHAPTER 15

Generalized method of moments (GMM) Readings: Hamilton Ch. 14∗ ; Davidson and MacKinnon, Ch. 17 (see pg. 587 for refs. to applications); Newey and McFadden (1994), “Large Sample Estimation and Hypothesis Testing,” in Handbook of Econometrics, Vol. 4, Ch. 36.

15.1. Definition We’ve already seen one example of GMM in the introduction, based upon the χ2 distribution. Consider the following example based upon the t-distribution. The density function of a t-distributed r.v. Yt is 0

fYt (yt , θ ) =

Γ θ0 + 1 /2

(πθ0 )

1/2

Γ (θ0 /2)

1 + yt2 /θ0

−(θ0 +1)/2

Given an iid sample of size n, one could estimate θ0 by maximizing the log-likelihood function θˆ ≡ arg max ln Ln (θ) = Θ

n

∑ ln fYt (yt , θ)

t=1

• This approach is attractive since ML estimators are asymptotically efficient. This is because the ML estimator uses all of the available information (e.g., the distribution is fully specified up to a parameter). Recalling that a distribution is completely characterized by its moments, the ML estimator is interpretable as a GMM estimator that uses all of the moments. The method of moments estimator uses only K moments to estimate a K− dimensional 295

15.1. DEFINITION

296

parameter. Since information is discarded, in general, by the MM estimator, efficiency is lost relative to the ML estimator. • Continuing with the example, a t-distributed r.v. with density fYt (yt , θ0 ) has mean zero and variance V (yt ) = θ0 / θ0 − 2 (for θ0 > 2).

• Using the notation introduced previously, define a moment condition m 1t (θ) = n n θ/ (θ − 2) − yt2 and m1 (θ) = 1/n ∑t=1 m1t (θ) = θ/ (θ − 2) − 1/n ∑t=1 yt2 . As before, when evaluated at the true parameter value θ0 , both Eθ0 m1t (θ0 ) = 0 and Eθ0 m1 (θ0 ) = 0.

ˆ ≡ 0 yields a MM estimator: • Choosing θˆ to set m1 (θ) θˆ =

(15.1.1)

2 1−

n ∑i y2i

This estimator is based on only one moment of the distribution - it uses less information than the ML estimator, so it is intuitively clear that the MM estimator will be inefficient relative to the ML estimator. • An alternative MM estimator could be based upon the fourth moment of the t-distribution. The fourth moment of a t-distributed r.v. is 2 3 θ0 4 µ4 ≡ E(yt ) = 0 , (θ − 2) (θ0 − 4) provided θ0 > 4. We can define a second moment condition m2 (θ) =

3 (θ)2 1 n − ∑ yt4 (θ − 2) (θ − 4) n t=1

ˆ ≡ 0. If you solve • A second, different MM estimator chooses θˆ to set m2 (θ) this you’ll see that the estimate is different from that in equation 15.1.1. This estimator isn’t efficient either, since it uses only one moment. A GMM estimator would use the two moment conditions together to estimate the single parameter. The

15.1. DEFINITION

297

GMM estimator is overidentified, which leads to an estimator which is efficient relative to the just identified MM estimators (more on efficiency later). • As before, set mn (θ) = (m1 (θ), m2 (θ))0 . The n subscript is used to indicate the

sample size. Note that m(θ0 ) = O p (n−1/2 ), since it is an average of centered

random variables, whereas m(θ) = O p (1), θ 6= θ0 , where expectations are

taken using the true distribution with parameter θ0 . This is the fundamental reason that GMM is consistent. • A GMM estimator requires defining a measure of distance, d (m(θ)). A popular choice (for reasons noted below) is to set d (m(θ)) = m0Wn m, and we minimize sn (θ) = m(θ)0Wn m(θ). We assume Wn converges to a finite positive definite matrix. • In general, assume we have g moment conditions, so m(θ) is a g -vector and W is a g × g matrix. For the purposes of this course, the following definition of the GMM estimator is sufficiently general: D EFINITION 24. The GMM estimator of the K -dimensional parameter vector n θ0 , θˆ ≡ arg minΘ sn (θ) ≡ mn (θ)0Wn mn (θ), where mn (θ) = n1 ∑t=1 mt (θ) is a g-vector,

g ≥ K, with Eθ m(θ) = 0, and Wn converges almost surely to a finite g × g symmetric positive definite matrix W∞ . What’s the reason for using GMM if MLE is asymptotically efficient? • Robustness: GMM is based upon a limited set of moment conditions. For consistency, only these moment conditions need to be correctly specified, whereas MLE in effect requires correct specification of every conceivable moment condition. GMM is robust with respect to distributional misspecification. The price for robustness is loss of efficiency with respect to the MLE

15.2. CONSISTENCY

298

estimator. Keep in mind that the true distribution is not known so if we erroneously specify a distribution and estimate by MLE, the estimator will be inconsistent in general (not always). – Feasibility: in some cases the MLE estimator is not available, because we are not able to deduce the likelihood function. More on this in the section on simulation-based estimation. The GMM estimator may still be feasible even though MLE is not possible.

15.2. Consistency We simply assume that the assumptions of Theorem 19 hold, so the GMM estimator is strongly consistent. The only assumption that warrants additional comments is that of identification. In Theorem 19, the third assumption reads: (c) Identification: s∞ (·) has a unique global maximum at θ0 , i.e., s∞ (θ0 ) > s∞ (θ), ∀θ 6= θ0 . Taking the case of a quadratic objective function sn (θ) = mn (θ)0Wn mn (θ), first consider mn (θ). a.s.

• Applying a uniform law of large numbers, we get mn (θ) → m∞ (θ).

• Since Eθ0 mn (θ0 ) = 0 by assumption, m∞ (θ0 ) = 0.

• Since s∞ (θ0 ) = m∞ (θ0 )0W∞ m∞ (θ0 ) = 0, in order for asymptotic identification,

we need that m∞ (θ) 6= 0 for θ 6= θ0 , for at least some element of the vector. a.s.

This and the assumption that Wn → W∞ , a finite positive g × g definite g × g

matrix guarantee that θ0 is asymptotically identified.

• Note that asymptotic identification does not rule out the possibility of lack of identification for a given data set - there may be multiple minimizing solutions in finite samples.


299

15.3. Asymptotic normality We also simply assume that the conditions of Theorem 22 hold, so we will have asymptotic normality. However, we do need to find the structure of the asymptotic variance-covariance matrix of the estimator. From Theorem 22, we have √

d n θˆ − θ0 → N 0, J∞ (θ0 )−1 I∞ (θ0 )J∞ (θ0 )−1

where J∞ (θ0 ) is the almost sure limit of

√ ∂ ∂2 0 0 ∂θ∂θ0 sn (θ) and I∞ (θ ) = limn→∞ Var n ∂θ sn (θ ).

We need to determine the form of these matrices given the objective function s n (θ) = mn (θ)0Wn mn (θ). Now using the product rule from the introduction, ∂ ∂ 0 sn (θ) = 2 m (θ) Wn mn (θ) ∂θ ∂θ n Define the K × g matrix Dn (θ) ≡

∂ 0 m (θ) , ∂θ n

so: ∂ s(θ) = 2D(θ)W m (θ) . ∂θ

(15.3.1)

(Note that sn (θ), Dn (θ), Wn and mn (θ) all depend on the sample size n, but it is omitted to unclutter the notation). To take second derivatives, let Di be the i− th row of D(θ). Using the product rule, ∂2 ∂ s(θ) = 2Di (θ)Wn m (θ) ∂θ0 ∂θi ∂θ0 ∂ 0 0 0 = 2DiW D + 2m W D ∂θ0 i


When evaluating the term

at θ0 , assume that

∂ 0 ∂θ0 D(θ)i

∂ 2m(θ) W D(θ)0i ∂θ0 0

300

satisfies a LLN, so that it converges almost surely to a finite

limit. In this case, we have ∂ 0 0 a.s. D(θ )i → 0, 2m(θ ) W ∂θ0 0 0

a.s.

since m(θ0 ) = o p (1), W → W∞ . Stacking these results over the K rows of D, we get lim

∂2 sn (θ0 ) = J∞ (θ0 ) = 2D∞W∞ D0∞ , a.s., ∂θ∂θ0

where we define lim D = D∞ , a.s., and limW = W∞ , a.s. (we assume a LLN holds). With regard to I∞ (θ0 ), following equation 15.3.1, and noting that the scores have mean zero at θ0 (since E m(θ0 ) = 0 by assumption), we have I∞ (θ0 ) =

= =

√ ∂ lim Var n sn (θ0 ) n→∞ ∂θ lim E 4nDnWn m(θ0 )m(θ)0Wn D0n

n→∞

lim E 4DnWn

n→∞

√

nm(θ0 )

√ nm(θ)0 Wn D0n

Now, given that m(θ0 ) is an average of centered (mean-zero) quantities, it is reasonable √ to expect a CLT to apply, after multiplication by n. Assuming this, √ d nm(θ0 ) → N(0, Ω∞ ), where Ω∞ = lim E nm(θ0 )m(θ0 )0 . n→∞

15.4. CHOOSING THE WEIGHTING MATRIX

301

Using this, and the last equation, we get I∞ (θ0 ) = 4D∞W∞ Ω∞W∞ D0∞

Using these results, the asymptotic normality theorem gives us √

d h −1 −1 i n θˆ − θ0 → N 0, D∞W∞ D0∞ D∞W∞ Ω∞W∞ D0∞ D∞W∞ D0∞ ,

the asymptotic distribution of the GMM estimator for arbitrary weighting matrix Wn . Note that for J∞ to be positive definite, D∞ must have full row rank, ρ(D∞ ) = k.

15.4. Choosing the weighting matrix W is a weighting matrix, which determines the relative importance of violations of the individual moment conditions. For example, if we are much more sure of the first moment condition, which is based upon the variance, than of the second, which is based upon the fourth moment, we could set 

W =

a 0 0 b

 

with a much larger than b. In this case, errors in the second moment condition have less weight in the objective function. • Since moments are not independent, in general, we should expect that there be a correlation between the moment conditions, so it may not be desirable to set the off-diagonal elements to 0. W may be a random, data dependent matrix.

15.4. CHOOSING THE WEIGHTING MATRIX

302

• We have already seen that the choice of W will influence the asymptotic distribution of the GMM estimator. Since the GMM estimator is already inefficient w.r.t. MLE, we might like to choose the W matrix to make the GMM estimator efficient within the class of GMM estimators defined by mn (θ). • To provide a little intuition, consider the linear model y = x0 β + ε, where ε ∼ N(0, Ω). That is, he have heteroscedasticity and autocorrelation.

• Let P be the Cholesky factorization of Ω−1 , e.g, P0 P = Ω−1 .

• Then the model Py = PXβ + Pε satisfies the classical assumptions of homoscedasticity and nonautocorrelation, since V (Pε) = PV (ε)P0 = PΩP0 = P(P0 P)−1 P0 = PP−1 (P0 )−1 P0 = In . (Note: we use (AB)−1 = B−1 A−1 for A, B both nonsingular). This means that the transformed model is efficient. • The OLS estimator of the model Py = PXβ + Pε minimizes the objective function (y−Xβ)0 Ω−1 (y−Xβ). Interpreting (y − Xβ) = ε(β) as moment con-

ditions (note that they do have zero expectation when evaluated at β 0 ), the

optimal weighting matrix is seen to be the inverse of the covariance matrix of the moment conditions. This result carries over to GMM estimation. (Note: this presentation of GLS is not a GMM estimator, because the number of moment conditions here is equal to the sample size, n. Later we’ll see that GLS can be put into the GMM framework defined above). T HEOREM 25. If θˆ is a GMM estimator that minimizes mn (θ)0Wn mn (θ), the asa.s ymptotic variance of θˆ will be minimized by choosing Wn so that Wn → W∞ = Ω−1 ∞ , where Ω∞ = limn→∞ E nm(θ0 )m(θ0 )0 .

Proof: For W∞ = Ω−1 ∞ , the asymptotic variance D∞W∞ D0∞

−1

D∞W∞ Ω∞W∞ D0∞ D∞W∞ D0∞

−1

15.4. CHOOSING THE WEIGHTING MATRIX 0 simplifies to D∞ Ω−1 ∞ D∞

−1

303

. Now, for any choice such that W∞ 6= Ω−1 ∞ , consider the

difference of the inverses of the variances when W = Ω−1 versus when W is some arbitrary positive definite matrix: 0 0 0 −1 0 D∞ Ω−1 D − D W D D W Ω W D D W D ∞ ∞ ∞ ∞ ∞ ∞ ∞ ∞ ∞ ∞ ∞ ∞ ∞ i h −1 −1/2 1/2 D∞W∞ Ω1/2 Ω−1/2 D0∞ = D ∞ Ω∞ I − Ω∞ W∞ D0∞ D∞W∞ Ω∞W∞ D0∞ ∞ ∞

as can be verified by multiplication. The term in brackets is idempotent, which is also easy to check by multiplication, and is therefore positive semidefinite. A quadratic form in a positive semidefinite matrix is also positive semidefinite. The difference of the inverses of the variances is positive semidefinite, which implies that the difference of the variances is negative semidefinite, which proves the theorem. The result

√

(15.4.1) allows us to treat

i d h 0 −1 n θˆ − θ0 → N 0, D∞ Ω−1 D ∞ ∞ 0 D∞ Ω−1 ∞ D∞ θˆ ≈ N θ0 , n

−1 !

,

where the ≈ means ”approximately distributed as.” To operationalize this we need

estimators of D∞ and Ω∞ .

c∞ is simply • The obvious estimator of D ˆ assuming that consistency of θ,

∂ 0 ∂θ mn

∂ 0 ∂θ mn

is continuous in θ. Stochastic equicon-

tinuity results can give us this result even if turn to estimation of Ω∞ .

θˆ , which is consistent by the

∂ 0 ∂θ mn

is not continuous. We now

15.5. ESTIMATION OF THE VARIANCE-COVARIANCE MATRIX

304

15.5. Estimation of the variance-covariance matrix (See Hamilton Ch. 10, pp. 261-2 and 280-84)∗ . In the case that we wish to use the optimal weighting matrix, we need an estimate √ of Ω∞ , the limiting variance-covariance matrix of nmn (θ0 ). While one could estimate Ω∞ parametrically, we in general have little information upon which to base a parametric specification. In general, we expect that: 0 ) 6= 0). Note that this autocovari• mt will be autocorrelated (Γts = E (mt mt−s

ance will not depend on t if the moment conditions are covariance stationary. • contemporaneously correlated, since the individual moment conditions will not in general be independent of one another (E (mit m jt ) 6= 0).

• and have different variances (E (m2it ) = σ2it ).

Since we need to estimate so many components if we are to take the parametric approach, it is unlikely that we would arrive at a correct parametric specification. For this reason, research has focused on consistent nonparametric estimators of Ω ∞ . Henceforth we assume that mt is covariance stationary (the covariance between mt and mt−s does not depend on t). Define the v −th autocovariance of the moment condi-

0 ). Note that E (m m0 ) = Γ0 . Recall that m and m are functions of tions Γv = E (mt mt−s t t t+s v

ˆ θ, so for now assume that we have some consistent estimator of θ 0 , so that mˆ t = mt (θ). Now Ωn

!# ! " n n 1/n ∑ mt0 = E nm(θ0 )m(θ0 )0 = E n 1/n ∑ mt "

= E 1/n = Γ0 +

n

∑ mt

t=1

!

n

∑ mt0

t=1

!#

t=1

t=1

n−2 n−1 1 Γ1 + Γ01 + Γ2 + Γ02 · · · + Γn−1 + Γ0n−1 n n n


305

A natural, consistent estimator of Γv is Γbv = 1/n

n

∑

t=v+1

0 mˆ t mˆ t−v .

(you might use n − v in the denominator instead). So, a natural, but inconsistent,

estimator of Ω∞ would be

n − 1 c c0 n − 2 c c0 0 d c d ˆ Ω = Γ0 + Γ1 + Γ1 + Γ2 + Γ2 + · · · + Γn−1 + Γn−1 n n n−1 c0 + ∑ n − v Γbv + Γb0v . = Γ v=1 n

This estimator is inconsistent in general, since the number of parameters to estimate is more than the number of observations, and increases more rapidly than n, so information does not build up as n → ∞.

On the other hand, supposing that Γv tends to zero sufficiently rapidly as v tends to

∞, a modified estimator

q(n) ˆ c Ω = Γ0 + ∑ Γbv + Γb0v , v=1

p

where q(n) → ∞ as n → ∞ will be consistent, provided q(n) grows sufficiently slowly. The term

n−v n

can be dropped because q(n) must be o p (n). This allows information

to accumulate at a rate that satisfies a LLN. A disadvantage of this estimator is that it may not be positive definite. This could cause one to calculate a negative χ 2 statistic, for example!

ˆ requires an estimate of m(θ0 ), which in turn requires • Note: the formula for Ω an estimate of θ, which is based upon an estimate of Ω! The solution to this circularity is to set the weighting matrix W arbitrarily (for example to an identity matrix), obtain a first consistent but inefficient estimate of θ0 , then


306

ˆ then re-estimate θ0 . The process can be iterated use this estimate to form Ω, ˆ nor θˆ change appreciably between iterations. until neither Ω 15.5.1. Newey-West covariance estimator. The Newey-West estimator (Econometrica, 1987) solves the problem of possible nonpositive definiteness of the above estimator. Their estimator is q(n) c ˆ Ω = Γ0 + ∑ 1 − v=1

v Γbv + Γb0v . q+1

This estimator is p.d. by construction. The condition for consistency is that n −1/4 q → 0. Note that this is a very slow rate of growth for q. This estimator is nonparametric we’ve placed no parametric restrictions on the form of Ω. It is an example of a kernel estimator. In a more recent paper, Newey and West (Review of Economic Studies, 1994) use pre-whitening before applying the kernel estimator. The idea is to fit a VAR model to the moment conditions. It is expected that the residuals of the VAR model will be more nearly white noise, so that the Newey-West covariance estimator might perform better with short lag lengths.. The VAR model is mˆ t = Θ1 mˆ t−1 + · · · + Θ p mˆ t−p + ut This is estimated, giving the residuals uˆt . Then the Newey-West covariance estimator is applied to these pre-whitened residuals, and the covariance Ω is estimated combining the fitted VAR c1 mˆ t−1 + · · · + Θ cp mˆ t−p c m ˆt = Θ

with the kernel estimate of the covariance of the ut . See Newey-West for details. • I have a program that does this if you’re interested.

15.6. ESTIMATION USING CONDITIONAL MOMENTS

307

15.6. Estimation using conditional moments So far, the moment conditions have been presented as unconditional expectations. One common way of defining unconditional moment conditions is based upon conditional moment conditions. Suppose that a random variable Y has zero expectation conditional on the random variable X EY |X Y =

Z

Y f (Y |X )dY = 0

Then the unconditional expectation of the product of Y and a function g(X ) of X is also zero. The unconditional expectation is E Y g(X ) =

Z Z X

Y

Y g(X ) f (Y, X )dY dX .

This can be factored into a conditional expectation and an expectation w.r.t. the marginal density of X : E Y g(X ) =

Z Z X

Y

Y g(X ) f (Y |X )dY

f (X )dX .

Since g(X ) doesn’t depend on Y it can be pulled out of the integral E Y g(X ) =

Z Z X

Y

Y f (Y |X )dY g(X ) f (X )dX .

But the term in parentheses on the rhs is zero by assumption, so E Y g(X ) = 0

as claimed. This is important econometrically, since models often imply restrictions on conditional moments. Suppose a model tells us that the function K(yt , xt ) has expectation,


308

conditional on the information set It , equal to k(xt , θ), Eθ K(yt , xt )|It = k(xt , θ).

• For example, in the context of the classical linear model yt = xt0 β + εt , we can set K(yt , xt ) = yt so that k(xt , θ) = xt0 β.

With this, the function ht (θ) = K(yt , xt ) − k(xt , θ) has conditional expectation equal to zero Eθ ht (θ)|It = 0.

This is a scalar moment condition, which isn’t sufficient to identify a K -dimensional parameter θ (K > 1). However, the above result allows us to form various unconditional expectations mt (θ) = Z(wt )ht (θ) where Z(wt ) is a g × 1-vector valued function of wt and wt is a set of variables drawn from the information set It . The Z(wt ) are instrumental variables. We now have g moment conditions, so as long as g > K the necessary condition for identification holds.


One can form the n × g matrix   Z1 (w1 ) Z2 (w1 ) · · ·   Z1 (w2 ) Z2 (w2 )  Zn =  .  ..   Z1 (wn ) Z2 (wn ) · · ·   0  Z1     Z0   2  =         Zn0

With this we can form the g moment conditions    1 0  Z  mn (θ) = n n  

309



Zg (w1 )   Zg (w2 )    ..  .   Zg (wn )



h1 (θ)   h2 (θ)    ..  .   hn (θ)

1 0 Z hn (θ) n n 1 n = ∑ Zt ht (θ) n t=1

=

1 n = ∑ mt (θ) n t=1 where Z(t,·) is the t th row of Zn . This fits the previous treatment. An interesting question that arises is how one should choose the instrumental variables Z(wt ) to achieve maximum efficiency.


Note that with this choice of moment conditions, we have that Dn ≡

310 ∂ 0 ∂θ m (θ)

(a

K × g matrix) is 0 ∂ 1 0 Zn hn (θ) ∂θ n 1 ∂ 0 = h (θ) Zn n ∂θ n

Dn (θ) =

which we can define to be 1 Dn (θ) = Hn Zn . n where Hn is a K × n matrix that has the derivatives of the individual moment conditions as its columns. Likewise, define the var-cov. of the moment conditions Ωn = E nmn (θ0 )mn (θ0 )0 1 0 0 0 0 = E Zn hn (θ )hn (θ ) Zn n 1 0 0 0 0 = Zn E hn (θ )hn (θ ) Zn n ≡ Zn0

Φn Zn n

where we have defined Φn = Varhn (θ0 ). Note that the dimension of this matrix is growing with the sample size, so it is not consistently estimable without additional assumptions. The asymptotic normality theorem above says that the GMM estimator using the optimal weighting matrix is distributed as √

d n θˆ − θ0 → N(0,V∞ )


311

where (15.6.1)

V∞ = lim

n→∞

Hn Zn n

Zn0 Φn Zn n

−1

Zn0 Hn0 n

!−1

.

Using an argument similar to that used to prove that Ω−1 ∞ is the efficient weighting matrix, we can show that putting 0 Zn = Φ−1 n Hn

causes the above var-cov matrix to simplify to (15.6.2)

V∞ = lim

n→∞

0 Hn Φ−1 n Hn n

−1

.

and furthermore, this matrix is smaller that the limiting var-cov for any other choice of instrumental variables. (To prove this, examine the difference of the inverses of the var-cov matrices with the optimal intruments and with non-optimal instruments. As above, you can show that the difference is positive semi-definite). • Note that both Hn , which we should write more properly as Hn (θ0 ), since it depends on θ0 , and Φ must be consistently estimated to apply this.

• Usually, estimation of Hn is straightforward - one just uses b = ∂ h0n θ˜ , H ∂θ

where θ˜ is some initial consistent estimator based on non-optimal instruments. • Estimation of Φn may not be possible. It is an n × n matrix, so it has more unique elements than n, the sample size, so without restrictions on the parameters it can’t be estimated consistently. Basically, you need to provide a parametric specification of the covariances of the ht (θ) in order to be able

15.8. A SPECIFICATION TEST

312

to use optimal instruments. A solution is to approximate this matrix parametrically to define the instruments. Note that the simplified var-cov matrix in equation 15.6.2 will not apply if approximately optimal instruments are used - it will be necessary to use an estimator based upon equation 15.6.1, where the term

Zn0 Φn Zn n

must be estimated consistently apart, for example by

the Newey-West procedure.

15.7. Estimation using dynamic moment conditions Note that dynamic moment conditions simplify the var-cov matrix, but are often harder to formulate. The will be added in future editions. For now, the Hansen application below is enough.

15.8. A specification test The first order conditions for minimization, using the an estimate of the optimal weighting matrix, are ∂ 0 ˆ ˆ −1 ∂ ˆ s(θ) = 2 mn θ Ω mn θˆ ≡ 0 ∂θ ∂θ

or

ˆ Ω ˆ −1 mn (θ) ˆ ≡0 D(θ) ˆ Consider a Taylor expansion of m(θ):

(15.8.1)

ˆ = mn (θ0 ) + D0n (θ0 ) θˆ − θ0 + o p (1). m(θ)


313

ˆ Ω ˆ −1 we obtain Multiplying by D(θ) ˆ Ω ˆ = D(θ) ˆ Ω ˆ Ω ˆ −1 m(θ) ˆ −1 mn (θ0 ) + D(θ) ˆ −1 D(θ0 )0 θˆ − θ0 + o p (1) D(θ)

ˆ tends to Ω∞ , we can write The lhs is zero, and since θˆ tends to θ0 and Ω 0 a −1 0 ˆ 0 D∞ Ω−1 ∞ mn (θ ) = −D∞ Ω∞ D∞ θ − θ

or √

a √ 0 −1 0 n θˆ − θ0 = − n D∞ Ω−1 D∞ Ω−1 ∞ D∞ ∞ mn (θ )

With this, and taking into account the original expansion (equation 15.8.1), we get √

a

ˆ = nm(θ)

√ √ 0 −1 0 nmn (θ0 ) − nD0∞ D∞ Ω−1 D∞ Ω−1 ∞ D∞ ∞ mn (θ ).

This last can be written as

√ a √ 1/2 0 −1 −1/2 ˆ = nm(θ) n Ω∞ − D0∞ D∞ Ω−1 D D Ω Ω−1/2 mn (θ0 ) ∞ ∞ ∞ ∞ ∞

Or

√ −1/2 a √ −1/2 −1/2 0 −1 0 −1 ˆ D∞ Ω∞ Ω−1/2 mn (θ0 ) nΩ∞ m(θ) = n Ig − Ω∞ D∞ D∞ Ω∞ D∞ ∞

Now

√ −1/2 d nΩ∞ mn (θ0 ) → N(0, Ig ) and one can easily verify that −1/2 −1/2 0 0 −1 D Ω P = Ig − Ω∞ D∞ D∞ Ω−1 D ∞ ∞ ∞ ∞


314

is idempotent of rank g − K, (recall that the rank of an idempotent matrix is equal to its trace) so √

−1/2 ˆ nΩ∞ m(θ)

0 √ d 2 −1/2 −1 ˆ = nm(θ) ˆ 0 Ω∞ ˆ → nΩ∞ m(θ) m(θ) χ (g − K)

ˆ converges to Ω∞ , we also have Since Ω d 2 ˆ 0Ω ˆ −1 m(θ) ˆ → nm(θ) χ (g − K)

or d 2 ˆ → n · sn (θ) χ (g − K)

supposing the model is correctly specified. This is a convenient test since we just multiply the optimized value of the objective function by n, and compare with a χ 2 (g − K) critical value. The test is a general test of whether or not the moments used to estimate are correctly specified. • This won’t work when the estimator is just identified. The f.o.c. are ˆ −1 m(θ) ˆ ≡ 0. Dθ sn (θ) = DΩ ˆ are square and invertible (at least But with exact identification, both D and Ω asymptotically, assuming that asymptotic normality hold), so ˆ ≡ 0. m(θ) So the moment conditions are zero regardless of the weighting matrix used. As such, we might as well use an identity matrix and save trouble. Also ˆ = 0, so the test breaks down. sn (θ) • A note: this sort of test often over-rejects in finite samples. One should be cautious in rejecting a model when this test rejects.

15.9. OTHER ESTIMATORS INTERPRETED AS GMM ESTIMATORS

315

15.9. Other estimators interpreted as GMM estimators 15.9.1. OLS with heteroscedasticity of unknown form. E XAMPLE 26. White’s heteroscedastic consistent varcov estimator for OLS. Suppose y = Xβ0 + ε, where ε ∼ N(0, Σ), Σ a diagonal matrix. • The typical approach is to parameterize Σ = Σ(σ), where σ is a finite dimensional parameter vector, and to estimate β and σ jointly (feasible GLS). This

will work well if the parameterization of Σ is correct. • If we’re not confident about parameterizing Σ, we can still estimate β consisˆ = (X0 X)−1 σˆ 2 tently by OLS. However, the typical covariance estimator V (β) will be biased and inconsistent, and will lead to invalid inferences. By exogeneity of the regressors xt (a K × 1 column vector) we have E(xt εt ) = 0,which suggests the moment condition mt (β) = xt yt − xt0 β .

In this case, we have exact identification ( K parameters and K moment conditions). We have m(β) = 1/n ∑ mt = 1/n ∑ xt yt − 1/n ∑ xt xt0 β. t

t

t

For any choice of W, m(β) will be identically zero at the minimum, due to exact identification. That is, since the number of moment conditions is identical to the number ˆ ≡ 0 regardless of W. There is no need to use the of parameters, the foc imply that m(β) “optimal” weighting matrix in this case, an identity matrix works just as well for the purpose of estimation. Therefore βˆ =

∑ t

xt xt0

−1

∑ xt yt = (X0X)−1X0y, t


316

which is the usual OLS estimator.

−1 b −1 D c∞ Ω c∞ 0 . Recall The GMM estimator of the asymptotic varcov matrix is D c∞ is simply ∂ m0 θˆ . In this case that D ∂θ c∞ = −1/n ∑ xt xt0 = −X0 X/n. D t

Recall that a possible estimator of Ω is

n−1 0 b b ˆ c Ω = Γ0 + ∑ Γv + Γv . v=1

This is in general inconsistent, but in the present case of nonautocorrelation, it simplifies to c0 ˆ =Γ Ω

which has a constant number of elements to estimate, so information will accumulate, and consistency obtains. In the present case b = Γ c0 = 1/n Ω = 1/n

"

n

= 1/n

n

t=1

!

yt − xt0 βˆ

∑ xt xt0 εˆt2

t=1

=

∑ mˆ t mˆ t0

∑ xt xt0

t=1

"

n

#

2

#

ˆ X0 EX n

where Eˆ is an n × n diagonal matrix with εˆ t2 in the position t,t.


317

Therefore, the GMM varcov. estimator, which is consistent, is ! )−1 ˆ −1 X0 X X0 X X0 EX − − n n n 0 −1 0 −1 0 ˆ X EX XX XX = n n n

√ Vˆ n βˆ − β =

(

This is the varcov estimator that White (1980) arrived at in an influential article. This estimator is consistent under heteroscedasticity of an unknown form. If there is autocorrelation, the Newey-West estimator can be used to estimate Ω - the rest is the same. 15.9.2. Weighted Least Squares. Consider the previous example of a linear model with heteroscedasticity of unknown form: y = Xβ0 + ε ε ∼ N(0, Σ) where Σ is a diagonal matrix. Now, suppose that the form of Σ is known, so that Σ(θ0 ) is a correct parametric specification (which may also depend upon X). In this case, the GLS estimator is β˜ = X0 Σ−1 X

−1

X0 Σ−1 y)

This estimator can be interpreted as the solution to the K moment conditions ˜ = 1/n m(β) ∑ t

xt xt0 ˜ xt yt − 1/n ∑ σt (θ0) β ≡ 0. σt (θ0 ) t

That is, the GLS estimator in this case has an obvious representation as a GMM estimator. With autocorrelation, the representation exists but it is a little more complicated. Nevertheless, the idea is the same. There are a few points:


318

• The (feasible) GLS estimator is known to be asymptotically efficient in the class of linear asymptotically unbiased estimators (Gauss-Markov). • This means that it is more efficient than the above example of OLS with White’s heteroscedastic consistent covariance, which is an alternative GMM estimator. • This means that the choice of the moment conditions is important to achieve efficiency.

15.9.3. 2SLS. Consider the linear model yt = zt0 β + εt , or y = Zβ + ε using the usual construction, where β is K × 1 and εt is i.i.d. Suppose that this equation is one of a system of simultaneous equations, so that zt contains both endogenous and exogenous variables. Suppose that xt is the vector of all exogenous and predetermined variables that are uncorrelated with εt (suppose that xt is r × 1). ˆ as the vector of predictions of Z when regressed upon X, e.g., Z ˆ = • Define Z X (X0 X)−1 X0 Z

ˆ = X X0 X Z

−1

X0 Z

ˆ is a linear combination of the exogenous variables x, zˆ t must be un• Since Z

correlated with ε. This suggests the K-dimensional moment condition mt (β) = zˆ t (yt − zt0 β) and so m(β) = 1/n ∑ zˆ t yt − zt0 β . t


319

• Since we have K parameters and K moment conditions, the GMM estimator will set m identically equal to zero, regardless of W, so we have βˆ =

∑ t

zˆ t zt0

−1

∑ (ˆzt yt ) = t

ˆ 0Z Z

−1

ˆ 0y Z

This is the standard formula for 2SLS. We use the exogenous variables and the reduced form predictions of the endogenous variables as instruments, and apply IV estimation. See Hamilton pp. 420-21 for the varcov formula (which is the standard formula for 2SLS), and for how to deal with εt heterogeneous and dependent (basically, just use the Newey-West or some other consistent estimator of Ω, and apply the usual formula). Note that εt dependent causes lagged endogenous variables to loose their status as legitimate instruments.

15.9.4. Nonlinear simultaneous equations. GMM provides a convenient way to estimate nonlinear systems of simultaneous equations. We have a system of equations of the form y1t = f1 (zt , θ01 ) + ε1t y2t = f2 (zt , θ02 ) + ε2t .. . yGt = fG (zt , θ0G ) + εGt , or in compact notation yt = f (zt , θ0 ) + εt , 00 0 00 where f (·) is a G -vector valued function, and θ0 = (θ00 1 , θ2 , · · · , θG ) .


320

We need to find an Ai × 1 vector of instruments xit , for each equation, that are

uncorrelated with εit . Typical instruments would be low order monomials in the ex ogenous variables in zt , with their lagged values. Then we can define the ∑G i=1 Ai × 1 orthogonality conditions



 (y1t − f1 (zt , θ1 )) x1t   (y2t − f2 (zt , θ2 )) x2t  mt (θ) =  ..  .   (yGt − fG (zt , θG )) xGt



    .   

• A note on identification: selection of instruments that ensure identification is a non-trivial problem. • A note on efficiency: the selected set of instruments has important effects on the efficiency of estimation. Unfortunately there is little theory offering guidance on what is the optimal set. More on this later.

15.9.5. Maximum likelihood. In the introduction we argued that ML will in general be more efficient than GMM since ML implicitly uses all of the moments of the distribution while GMM uses a limited number of moments. Actually, a distribution with P parameters can be uniquely characterized by P moment conditions. However, some sets of P moment conditions may contain more information than others, since the moment conditions could be highly correlated. A GMM estimator that chose an optimal set of P moment conditions would be fully efficient. Here we’ll see that the optimal moment conditions are simply the scores of the ML estimator. Let yt be a G -vector of variables, and let Yt = (y01 , y02 , ..., yt0 )0 . Then at time t, Yt−1 has been observed (refer to it as the information set, since we assume the conditioning


321

variables have been selected to take advantage of all useful information). The likelihood function is the joint density of the sample: L (θ) = f (y1 , y2 , ..., yn, θ)

which can be factored as L (θ) = f (yn |Yn−1 , θ) · f (Yn−1 , θ)

and we can repeat this to get L (θ) = f (yn |Yn−1 , θ) · f (yn−1 |Yn−2 , θ) · ... · f (y1).

The log-likelihood function is therefore ln L (θ) =

n

∑ ln f (yt |Yt−1 , θ).

t=1

Define mt (Yt , θ) ≡ Dθ ln f (yt |Yt−1 , θ) as the score of the t th observation. It can be shown that, under the regularity conditions, that the scores have conditional mean zero when evaluated at θ 0 (see notes to Introduction to Econometrics): E {mt (Yt , θ0 )|Yt−1 } = 0

so one could interpret these as moment conditions to use to define a just-identified GMM estimator ( if there are K parameters there are K score equations). The GMM estimator sets n

n

t=1

t=1

ˆ = 1/n ∑ Dθ ln f (yt |Yt−1 , θ) ˆ = 0, 1/n ∑ mt (Yt , θ)


322

which are precisely the first order conditions of MLE. Therefore, MLE can be inter−1 preted as a GMM estimator. The GMM varcov formula is V∞ = D∞ Ω−1 D0∞ . Consistent estimates of variance components are as follows

• D∞

n ˆ = 1/n ∑ D2 ln f (yt |Yt−1 , θ) ˆ c∞ = ∂ m(Yt , θ) D θ ∂θ0 t=1

• Ω

It is important to note that mt and mt−s , s > 0 are both conditionally and unconditionally uncorrelated. Conditional uncorrelation follows from the fact that mt−s is a function of Yt−s , which is in the information set at time t. Unconditional uncorrelation follows from the fact that conditional uncorrelation hold regardless of the realization of Yt−1 , so marginalizing with respect to Yt−1 preserves uncorrelation (see the section on ML estimation, above). The fact that the scores are serially uncorrelated implies that Ω can be estimated by the estimator of the 0th autocovariance of the moment conditions: n n b = 1/n ∑ mt (Yt , θ)m ˆ Dθ ln f (yt |Yt−1 , θ) ˆ 0 ˆ t (Yt , θ) ˆ 0 = 1/n ∑ Dθ ln f (yt |Yt−1 , θ) Ω t=1

t=1

Recall from study of ML estimation that the information matrix equality (equation ??) states that E

n 0 o = −E D2θ ln f (yt |Yt−1 , θ0 ) . Dθ ln f (yt |Yt−1 , θ0 ) Dθ ln f (yt |Yt−1 , θ0 )

This result implies the well known (and already seeen) result that we can estimate V∞ in any of three ways:

15.10. EXAMPLE: THE HAUSMAN TEST

• The sandwich version:  n  ˆ ×  ∑t=1 D2θ ln f (yt |Yt−1 , θ)    n 0 o−1 n c ˆ ˆ V = n × ∞ ∑t=1 Dθ ln f (yt |Yt−1 , θ) Dθ ln f (yt |Yt−1 , θ)     n  ˆ D2θ ln f (yt |Yt−1 , θ) ∑t=1

323

−1          

• or the inverse of the negative of the Hessian (since the middle and last term cancel, except for a minus sign): " #−1 n 2 ˆ c V , ∞ = −1/n ∑ Dθ ln f (yt |Yt−1 , θ) t=1

• or the inverse of the outer product of the gradient (since the middle and last cancel except for a minus sign, and the first term converges to minus the inverse of the middle term, which is still inside the overall inverse) )−1 ( n 0 ˆ ˆ c . V ∞ = 1/n ∑ Dθ ln f (yt |Yt−1 , θ) Dθ ln f (yt |Yt−1 , θ) t=1

This simplification is a special result for the MLE estimator - it doesn’t apply to GMM estimators in general. Asymptotically, if the model is correctly specified, all of these forms converge to the same limit. In small samples they will differ. In particular, there is evidence that the outer product of the gradient formula does not perform very well in small samples (see Davidson and MacKinnon, pg. 477). White’s Information matrix test (Econometrica, 1982) is based upon comparing the two ways to estimate the information matrix: outer product of gradient or negative of the Hessian. If they differ by too much, this is evidence of misspecification of the model. 15.10. Example: The Hausman Test This section discusses the Hausman test, which was originally presented in Hausman, J.A. (1978), Specification tests in econometrics, Econometrica, 46, 1251-71.


324

Consider the simple linear regression model yt = xt0 β + εt . We assume that the functional form and the choice of regressors is correct, but that the some of the regressors may be correlated with the error term, which as you know will produce inconsistency ˆ For example, this will be a problem if of β. • if some regressors are endogeneous • some regressors are measured with error

• lagged values of the dependent variable are used as regressors and εt is autocorrelated. To illustrate, the Octave program biased.m performs a Monte Carlo experiment where errors are correlated with regressors, and estimation is by OLS and IV. Figure 15.10.1 shows that the OLS estimator is quite biased, while Figure 15.10.2 shows that the IV estimator is on average much closer to the true value. If you play with the program, increasing the sample size, you can see evidence that the OLS estimator is asymptotically biased, while the IV estimator is consistent. We have seen that inconsistent and the consistent estimators converge to different probability limits. This is the idea behind the Hausman test - a pair of consistent estimators converge to the same probability limit, while if one is consistent and the other is not they converge to different limits. If we accept that one is consistent (e.g., the IV estimator), but we are doubting if the other is consistent (e.g., the OLS estimator), we might try to check if the difference between the estimators is significantly different from zero. • If we’re doubting about the consistency of OLS (or QML, etc.), why should we be interested in testing - why not just use the IV estimator? Because the OLS estimator is more efficient when the regressors are exogenous and the other classical assumptions (including normality of the errors) hold. When we have a more efficient estimator that relies on stronger assumptions (such


325

F IGURE 15.10.1. OLS OLS estimates 0.14

line 1

0.12

0.1

0.08

0.06

0.04

0.02

0 2.26

2.28

2.3

2.32

2.34

2.36

2.38

2.4

F IGURE 15.10.2. IV IV estimates 0.16

line 1

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0 1.85

1.9

1.95

2

2.05

2.1

2.15


326

as exogeneity) than the IV estimator, we might prefer to use it, unless we have evidence that the assumptions are false.

So, let’s consider the covariance between the MLE estimator θˆ (or any other fully ˜ Now, let’s recall some efficient estimator) and some other CAN estimator, say θ. results from MLE. Equation 4.4.1 is: √ Equation 4.6.2 is

a.s. √ n θˆ − θ0 → −H∞ (θ0 )−1 ng(θ0 ). H∞ (θ) = −I∞ (θ).

Combining these two equations, we get √

a.s. √ n θˆ − θ0 → I∞ (θ0 )−1 ng(θ0 ).

Also, equation 4.7.1 tells us that the asymptotic covariance between any CAN estimator and the MLE score vector is   √   ˜ ˜ V (θ) IK n θ−θ . = ∞ V∞  √ IK I∞ (θ) ng(θ)

Now, consider  

IK 0K

 √  √   ˜ −θ θ n θ˜ − θ n    a.s. → √ . √ −1 ˆ I∞ (θ) ng(θ) n θ−θ 0K


327

The asymptotic covariance of this is  √ n V∞  √ n

     ˜ I 0K V (θ) IK I 0K  =  K  ∞  K  −1 −1 ˆθ − θ 0K I∞ (θ) IK I∞ (θ) 0K I∞ (θ)   ˜ V∞ (θ) I∞ (θ)−1 , =  −1 −1 I∞ (θ) I∞ (θ) θ˜ − θ

which, for clarity in what follows, we might write as

 √    ˜ n θ˜ − θ V∞ (θ) I∞ (θ)−1   . V∞  √ = −1 ˆ ˆ n θ−θ I∞ (θ) V∞ (θ)

So, the asymptotic covariance between the MLE and any other CAN estimator is equal to the MLE asymptotic variance (the inverse of the information matrix). Now, suppose we with to test whether the the two estimators are in fact both converging to θ0 , versus the alternative hypothesis that the ”MLE” estimator is not in fact consistent (the consistency of θ˜ is a maintained hypothesis). Under the null hypothesis that they are, we have h

IK −IK

i

 √  ˜ n θ − θ0 √ ˜ ˆ   = n θ−θ , √ ˆ n θ − θ0

will be asymptotically normally distributed as

So,

d √ ˜ −V∞ (θ) ˆ . n θ˜ − θˆ → N 0,V∞ (θ) n θ˜ − θˆ

0

˜ −V∞ (θ) ˆ V∞ (θ)

−1

d θ˜ − θˆ → χ2 (ρ),


328

where ρ is the rank of the difference of the asymptotic variances. A statistic that has the same asymptotic distribution is θ˜ − θˆ

0

˜ − Vˆ (θ) ˆ Vˆ (θ)

−1

d θ˜ − θˆ → χ2 (ρ).

This is the Hausman test statistic, in its original form. The reason that this test has power under the alternative hypothesis is that in that case the ”MLE” estimator will not be consistent, and will converge to θA , say, where θA 6= θ0 . Then the mean of the √ asymptotic distribution of vector n θ˜ − θˆ will be θ0 − θA , a non-zero vector, so the

test statistic will eventually reject, regardless of how small a significance level is used.

• Note: if the test is based on a sub-vector of the entire parameter vector of the MLE, it is possible that the inconsistency of the MLE will not show up in the portion of the vector that has been used. If this is the case, the test may not have power to detect the inconsistency. This may occur, for example, when the consistent but inefficient estimator is not identified for all the parameters of the model. Some things to note: • The rank, ρ, of the difference of the asymptotic variances is often less than the dimension of the matrices, and it may be difficult to determine what the true rank is. If the true rank is lower than what is taken to be true, the test will be biased against rejection of the null hypothesis. The contrary holds if we underestimate the rank. • A solution to this problem is to use a rank 1 test, by comparing only a single coefficient. For example, if a variable is suspected of possibly being endogenous, that variable’s coefficients may be compared.


329

• This simple formula only holds when the estimator that is being tested for consistency is fully efficient under the null hypothesis. This means that it must be a ML estimator or a fully efficient estimator that has the same asymptotic distribution as the ML estimator. This is quite restrictive since modern estimators such as GMM and QML are not in general fully efficient.

Following up on this last point, let’s think of two not necessarily efficient estimators, θˆ 1 and θˆ 2 , where one is assumed to be consistent, but the other may not be. We assume for expositional simplicity that both θˆ 1 and θˆ 2 belong to the same parameter space, and that they can be expressed as generalized method of moments (GMM) estimators. The estimators are defined (suppressing the dependence upon data) by θˆ i = arg min mi (θi )0 Wi mi (θi ) θi ∈Θ

where mi (θi ) is a gi ×1 vector of moment conditions, and Wi is a gi ×gi positive definite weighting matrix, i = 1, 2. Consider the omnibus GMM estimator (15.10.1)

θˆ 1 , θˆ 2 = arg min

Θ×Θ

h

m1 (θ1 )0 m2 (θ2 )0

i

 

W1

0(g1 ×g2 )

0(g2 ×g1 )

W2

 

m1 (θ1 ) m2 (θ2 )

Suppose that the asymptotic covariance of the omnibus moment vector is (15.10.2)



.

   √ m1 (θ1 )   Σ = lim Var n n→∞  m2 (θ2 )    Σ1 Σ12 . ≡  · Σ2

The standard Hausman test is equivalent to a Wald test of the equality of θ 1 and θ2 (or subvectors of the two) applied to the omnibus GMM estimator, but with the covariance

15.11. APPLICATION: NONLINEAR RATIONAL EXPECTATIONS

330

of the moment conditions estimated as   c Σ1 0(g1 ×g2 ) b= . Σ c 0(g2 ×g1 ) Σ2

While this is clearly an inconsistent estimator in general, the omitted Σ12 term cancels

out of the test statistic when one of the estimators is asymptotically efficient, as we have seen above, and thus it need not be estimated. The general solution when neither of the estimators is efficient is clear: the entire Σ matrix must be estimated consistently, since the Σ12 term will not cancel out. Methods for consistently estimating the asymptotic covariance of a vector of moment conditions are well-known, e.g., the Newey-West estimator discussed previously. The Hausman test using a proper estimator of the overall covariance matrix will now have an asymptotic χ2 distribution when neither estimator is efficient. However, the test suffers from a loss of power due to the fact that the omnibus GMM estimator of equation 15.10.1 is defined using an inefficient weight matrix. A new test can be defined by using an alternative omnibus GMM estimator (15.10.3)

θˆ 1 , θˆ 2 = arg min

Θ×Θ

h

m1 (θ1 )0

  i −1 m1 (θ1 ) e  , Σ m2 (θ2 )0 m2 (θ2 )

e is a consistent estimator of the overall covariance matrix Σ of equation 15.10.2. where Σ

By standard arguments, this is a more efficient estimator than that defined by equation 15.10.1, so the Wald test using this alternative is more powerful. See my article in Applied Economics, 2004, for more details, including simulation results.

15.11. Application: Nonlinear rational expectations Readings: Hansen and Singleton, 1982∗ ; Tauchen, 1986


331

Though GMM estimation has many applications, application to rational expectations models is elegant, since theory directly suggests the moment conditions. Hansen and Singleton’s 1982 paper is also a classic worth studying in itself. Though I strongly recommend reading the paper, I’ll use a simplified model with similar notation to Hamilton’s. We assume a representative consumer maximizes expected discounted utility over an infinite horizon. Utility is temporally additive, and the expected utility hypothesis ∞ holds. The future consumption stream is the stochastic sequence {ct }t=0 . The objec-

tive function at time t is the discounted expected utility (15.11.1)

∞

∑ βsE (u(ct+s)|It ) .

s=0

• The parameter β is between 0 and 1, and reflects discounting. • It is the information set at time t, and includes the all realizations of random variables indexed t and earlier. • The choice variable is ct - current consumption, which is constained to be less than or equal to current wealth wt . • Suppose the consumer can invest in a risky asset. A dollar invested in the asset yields a gross return (1 + rt+1 ) =

pt+1 + dt+1 pt

where pt is the price and dt is the dividend in period t. The price of ct is normalized to 1. • Current wealth wt = (1 + rt )it−1 , where it−1 is investment in period t − 1. So the problem is to allocate current wealth between current consumption and investment to finance future consumption: wt = ct + it .


332

• Future net rates of return rt+s , s > 0 are not known in period t: the asset is risky.

A partial set of necessary conditions for utility maximization have the form: (15.11.2)

u0 (ct ) = βE (1 + rt+1 ) u0 (ct+1 )|It .

To see that the condition is necessary, suppose that the lhs < rhs. Then by reducing current consumption marginally would cause equation 15.11.1 to drop by u0 (ct ), since there is no discounting of the current period. At the same time, the marginal reduction in consumption finances investment, which has gross return (1 + rt+1 ) , which could finance consumption in period t + 1. This increase in consumption would cause the objective function to increase by βE {(1 + rt+1 ) u0 (ct+1 )|It } . Therefore, unless the condition holds, the expected discounted utility function is not maximized.

• To use this we need to choose the functional form of utility. A constant relative risk aversion form is 1−γ

u(ct ) =

ct

−1 1−γ

where γ is the coefficient of relative risk aversion. With this form, −γ

u0 (ct ) = ct so the foc are

While it is true that

n o −γ −γ ct = βE (1 + rt+1 ) ct+1 |It

−γ

n

−γ

E ct − β (1 + rt+1 ) ct+1

o

|It = 0


333

so that we could use this to define moment conditions, it is unlikely that ct is stationary, even though it is in real terms, and our theory requires stationarity. To solve this, divide −γ

though by ct

(

ct+1 E 1-β (1 + rt+1 ) ct

−γ )!

|It = 0

(note that ct can be passed though the conditional expectation since ct is chosen based only upon information available in time t). Now

(

ct+1 1-β (1 + rt+1 ) ct

−γ )

is analogous to ht (θ) defined above: it’s a scalar moment condition. To get a vector of moment conditions we need some instruments. Suppose that zt is a vector of variables drawn from the information set It . We can use the necessary conditions to form the expressions

1 − β (1 + rt+1 )

• θ represents β and γ.

ct+1 ct

−γ

zt ≡ mt (θ)

• Therefore, the above expression may be interpreted as a moment condition which can be used for GMM estimation of the parameters θ0 .

Note that at time t, mt−s has been observed, and is therefore an element of the information set. By rational expectations, the autocovariances of the moment conditions other than Γ0 should be zero. The optimal weighting matrix is therefore the inverse of the variance of the moment conditions: Ω∞ = lim E nm(θ0 )m(θ0 )0


334

which can be consistently estimated by n

ˆ = 1/n ∑ mt (θ)m ˆ t (θ) ˆ 0 Ω t=1

As before, this estimate depends on an initial consistent estimate of θ, which can be obtained by setting the weighting matrix W arbitrarily (to an identity matrix, for exˆ we then minimize ample). After obtaining θ, ˆ −1 m(θ). s(θ) = m(θ)0 Ω This process can be iterated, e.g., use the new estimate to re-estimate Ω, use this to estimate θ0 , and repeat until the estimates don’t change.

• In principle, we could use a very large number of moment conditions in estimation, since any current or lagged variable could be used in xt . Since use of more moment conditions will lead to a more (asymptotically) efficient estimator, one might be tempted to use many instrumental variables. We will do a computer lab that will show that this may not be a good idea with finite samples. This issue has been studied using Monte Carlos (Tauchen, JBES, 1986). The reason for poor performance when using many instruments is that the estimate of Ω becomes very imprecise. • Empirical papers that use this approach often have serious problems in obtaining precise estimates of the parameters. Note that we are basing everything on a single parial first order condition. Probably this f.o.c. is simply not informative enough. Simulation-based estimation methods (discussed below) are one means of trying to use more informative moment conditions to estimate this sort of model.

15.12. EMPIRICAL EXAMPLE: A PORTFOLIO MODEL

335

15.12. Empirical example: a portfolio model The Octave program portfolio.m performs GMM estimation of a portfolio model, using the data file tauchen.data. The columns of this data file are c, p, and d in that order. There are 95 observations (source: Tauchen, JBES, 1986). As instruments we use lags of c and r, as well as a constant. For a single lag the estimation results are

MPITB extensions found

****************************************************** Example of GMM estimation of rational expectations model

GMM Estimation Results BFGS convergence: Normal convergence

Objective function value: 0.000014 Observations: 94

X^2 test

Value

df

p-value

0.001

1.000

0.971

estimate

st. err

t-stat

p-value

beta

0.915

0.009

97.271

0.000

gamma

0.569

0.319

1.783

0.075

******************************************************


336

For two lags the estimation results are MPITB extensions found

****************************************************** Example of GMM estimation of rational expectations model

GMM Estimation Results BFGS convergence: Normal convergence

Objective function value: 0.037882 Observations: 93

X^2 test

Value

df

p-value

3.523

3.000

0.318

estimate

st. err

t-stat

p-value

beta

0.857

0.024

35.636

0.000

gamma

-2.351

0.315

-7.462

0.000

******************************************************

Pretty clearly, the results are sensitive to the choice of instruments. Maybe there is some problem here: poor instruments, or possibly a conditional moment that is not very informative.


337

Exercises (1) Show how to cast the generalized IV estimator presented in section 11.4 as a GMM estimator. Identify what are the moment conditions, mt (θ), what is the form of the the matrix Dn , what is the efficient weight matrix, and show that the covariance matrix formula given previously corresponds to the GMM covariance matrix formula. (2) Using Octave, generate data from the logit dgp . Recall that E(yt |xt ) = p(xt , θ) = [1 + exp(−xt 0θ)]−1 . Consider the moment condtions (exactly iden-

tified) mt (θ) = [yt − p(xt , θ)]xt

(a) Estimate by GMM, using these moments. (b) Estimate by MLE. (c) The two estimators should coincide. Prove analytically that the estimators coicide. ˆ has a χ2 (g − ˆ 0Ω ˆ −1 m(θ) (3) Verify the missing steps needed to show that n · m(θ) K) distribution. That is, show that the monster matrix is idempotent and has trace equal to g − K. (4) For the portfolio example, experiment with the program using lags of 3 and 4 periods to define instruments (a) Iterate the estimation of θ = (β, γ) and Ω to convergence. (b) Comment on the results. Are the results sensitive to the set of instruments ˆ Are these good instruments? Are the ˆ as well as θ. used? (Look at Ω instruments highly correlated with one another?

CHAPTER 16

Quasi-ML Quasi-ML is the estimator one obtains when a misspecified probability model is used to calculate an ”ML” estimator. Given a sample of size n of a random vector y and a vector of conditioning variables x, suppose the joint density of Y = y1 . . . yn conditional on X = x1 . . . xn is a member of the parametric family pY (Y|X, ρ), ρ ∈ Ξ. The true joint density is as-

sociated with the vector ρ0 :

pY (Y|X, ρ0 ).

As long as the marginal density of X doesn’t depend on ρ0 , this conditional density fully characterizes the random characteristics of samples: i.e., it fully describes the probabilistically important features of the d.g.p. The likelihood function is just this density evaluated at other values ρ L(Y|X, ρ) = pY (Y|X, ρ), ρ ∈ Ξ. • Let Yt−1 =

y1 . . . yt−1

, Y0 = 0, and let Xt =

x1 . . . xt

The

likelihood function, taking into account possible dependence of observations, can be written as L(Y|X, ρ) =

n

∏ pt (yt |Yt−1, Xt , ρ)

t=1 n

≡

∏ pt (ρ)

t=1

338

16. QUASI-ML

339

• The average log-likelihood function is: 1 1 n sn (ρ) = ln L(Y|X, ρ) = ∑ ln pt (ρ) n n t=1 • Suppose that we do not have knowledge of the family of densities pt (ρ). Mistakenly, we may assume that the conditional density of yt is a member of the family ft (yt |Yt−1 , Xt , θ), θ ∈ Θ, where there is no θ0 such that ft (yt |Yt−1 , Xt , θ0 ) = pt (yt |Yt−1 , Xt , ρ0 ), ∀t (this is what we mean by “mis-

specified”). • This setup allows for heterogeneous time series data, with dynamic misspecification.

The QML estimator is the argument that maximizes the misspecified average log likelihood, which we refer to as the quasi-log likelihood function. This objective function is sn (θ) = ≡

1 n ln ft (yt |Yt−1 , Xt , θ0 ) ∑ n t=1 1 n ∑ ln ft (θ) n t=1

and the QML is θˆ n = arg max sn (θ) Θ

A SLLN for dependent sequences applies (we assume), so that 1 n sn (θ) → lim E ∑ ln ft (θ) ≡ s∞ (θ) n→∞ n t=1 a.s.

16.1. CONSISTENT ESTIMATION OF VARIANCE COMPONENTS

340

We assume that this can be strengthened to uniform convergence, a.s., following the previous arguments. The “pseudo-true” value of θ is the value that maximizes s(θ): ¯ θ0 = arg max s∞ (θ) Θ

Given assumptions so that theorem 19 is applicable, we obtain lim θˆ n = θ0 , a.s.

n→∞

• Applying the asymptotic normality theorem, √ where

d n θˆ − θ0 → N 0, J∞ (θ0 )−1 I∞ (θ0 )J∞ (θ0 )−1 J∞ (θ0 ) = lim E D2θ sn (θ0 ) n→∞

and √

I∞ (θ0 ) = lim Var nDθ sn (θ0 ). n→∞

• Note that asymptotic normality only requires that the additional assumptions

regarding J and I hold in a neighborhood of θ0 for J and at θ0 , for I , not

throughout Θ. In this sense, asymptotic normality is a local property. 16.1. Consistent Estimation of Variance Components Consistent estimation of J∞ (θ0 ) is straightforward. Assumption (b) of Theorem 22 implies that Jn (θˆ n ) =

1 n 2 1 n 2 a.s. ˆ ln f ( θ ) → lim E D ∑ θ t n n→∞ n ∑ Dθ ln ft (θ0) = J∞(θ0). n t=1 t=1

That is, just calculate the Hessian using the estimate θˆ n in place of θ0 . Consistent estimation of I∞ (θ0 ) is more difficult, and may be impossible.

16.1. CONSISTENT ESTIMATION OF VARIANCE COMPONENTS

341

• Notation: Let gt ≡ Dθ ft (θ0 ) We need to estimate I∞ (θ0 ) =

√ lim Var nDθ sn (θ0 )

n→∞

√ 1 n = lim Var n ∑ Dθ ln ft (θ0 ) n→∞ n t=1 n 1 = lim Var ∑ gt n→∞ n t=1 ! ( n 1 = lim E ∑ (gt − E gt ) n→∞ n t=1

n

!0 )

∑ (gt − E gt )

t=1

This is going to contain a term 1 n ∑ (E gt ) (E gt )0 n→∞ n t=1 lim

which will not tend to zero, in general. This term is not consistently estimable in general, since it requires calculating an expectation using the true density under the d.g.p., which is unknown. • There are important cases where I∞ (θ0 ) is consistently estimable. For example, suppose that the data come from a random sample (i.e., they are iid). This would be the case with cross sectional data, for example. (Note: under i.i.d. sampling, the joint distribution of (yt , xt ) is identical. This does not imply that the conditional density f (yt |xt ) is identical). • With random sampling, the limiting objective function is simply s∞ (θ0 ) = EX E0 ln f (y|x, θ0 ) where E0 means expectation of y|x and EX means expectation respect to the marginal density of x.

16.2. EXAMPLE: THE MEPS DATA

342

• By the requirement that the limiting objective function be maximized at θ 0 we have Dθ EX E0 ln f (y|x, θ0 ) = Dθ s∞ (θ0 ) = 0 • The dominated convergence theorem allows switching the order of expectation and differentiation, so Dθ EX E0 ln f (y|x, θ0 ) = EX E0 Dθ ln f (y|x, θ0 ) = 0 The CLT implies that 1 n d √ ∑ Dθ ln f (y|x, θ0 ) → N(0, I∞ (θ0 )). n t=1 That is, it’s not necessary to subtract the individual means, since they are zero. Given this, and due to independent observations, a consistent estimator is 1 n ˆ ˆ θ0 ln ft (θ) b I = ∑ Dθ ln ft (θ)D n t=1

This is an important case where consistent estimation of the covariance matrix is possible. Other cases exist, even for dynamically misspecified time series models. 16.2. Example: the MEPS Data To check the plausibility of the Poisson model for the MEPS data, we can compare the sample unconditional variance with the estimated unconditional variance according to the Poisson model: Vd (y) =

n λ ˆt ∑t=1 n .

Using the program PoissonVariance.m, for

OBDV and ERV, we get We see that even after conditioning, the overdispersion is not TABLE 1. Marginal Variances, Sample and Estimated (Poisson) OBDV ERV Sample 38.09 0.151 Estimated 3.28 0.086


343

captured in either case. There is huge problem with OBDV, and a significant problem with ERV. In both cases the Poisson model does not appear to be plausible. You can check this for the other use measures if you like. 16.2.1. Infinite mixture models: the negative binomial model. Reference: Cameron and Trivedi (1998) Regression analysis of count data, chapter 4. The two measures seem to exhibit extra-Poisson variation. To capture unobserved heterogeneity, a possibility is the random parameters approach. Consider the possibility that the constant term in a Poisson model were random: exp(−θ)θy fY (y|x, ε) = y! θ = exp(x0β + ε) = exp(x0β) exp(ε) = λν where λ = exp(x0β) and ν = exp(ε). Now ν captures the randomness in the constant. The problem is that we don’t observe ν, so we will need to marginalize it to get a usable density fY (y|x) =

Z ∞ exp[−θ]θy −∞

y!

fv (z)dz

This density can be used directly, perhaps using numerical integration to evaluate the likelihood function. In some cases, though, the integral will have an analytic solution. For example, if ν follows a certain one parameter gamma density, then (16.2.1)

Γ(y + ψ) fY (y|x, φ) = Γ(y + 1)Γ(ψ)

ψ ψ+λ

ψ

λ ψ+λ

y

where φ = (λ, ψ). ψ appears since it is the parameter of the gamma density. • For this density, E(y|x) = λ, which we have parameterized λ = exp(x 0 β)


344

• The variance depends upon how ψ is parameterized.

– If ψ = λ/α, where α > 0, then V (y|x) = λ + αλ. Note that λ is a function of x, so that the variance is too. This is referred to as the NB-I model. – If ψ = 1/α, where α > 0, then V (y|x) = λ + αλ2 . This is referred to as the NB-II model.

So both forms of the NB model allow for overdispersion, with the NB-II model allowing for a more radical form. Testing reduction of a NB model to a Poisson model cannot be done by testing α = 0 using standard Wald or LR procedures. The critical values need to be adjusted to account for the fact that α = 0 is on the boundary of the parameter space. Without getting into details, suppose that the data were in fact Poisson, so there is equidispersion and the true α = 0. Then about half the time the sample data will be underdispersed, and about half the time overdispersed. When the data is underdispersed, the MLE of α will be αˆ = 0. Thus, under the null, there will be a probability spike in the asymptotic p √ distribution of n(αˆ − α) = nαˆ at 0, so standard testing methods will not be valid. This program will do estimation using the NB model. Note how modelargs is used

to select a NB-I or NB-II density. Here are NB-I estimation results for OBDV:

MPITB extensions found OBDV ====================================================== BFGSMIN final results Used analytic gradient -----------------------------------------------------STRONG CONVERGENCE Function conv 1 Param conv 1 Gradient conv 1 ------------------------------------------------------


345

Objective function value 2.18573 Stepsize 0.0007 17 iterations -----------------------------------------------------param 1.0965 0.2551 0.2024 0.2289 0.1969 0.0769 0.0000 1.7146

gradient 0.0000 -0.0000 -0.0000 0.0000 0.0000 0.0000 -0.0000 -0.0000

change -0.0000 0.0000 0.0000 -0.0000 -0.0000 -0.0000 0.0000 0.0000

****************************************************** Negative Binomial model, MEPS 1996 full data set MLE Estimation Results BFGS convergence: Normal convergence Average Log-L: -2.185730 Observations: 4564 constant pub. ins. priv. ins. sex age edu inc alpha

estimate -0.523 0.765 0.451 0.458 0.016 0.027 0.000 5.555

st. err 0.104 0.054 0.049 0.034 0.001 0.007 0.000 0.296

t-stat -5.005 14.198 9.196 13.512 11.869 3.979 0.000 18.752

p-value 0.000 0.000 0.000 0.000 0.000 0.000 1.000 0.000

Information Criteria CAIC : 20026.7513 Avg. CAIC: 4.3880 BIC : 20018.7513 Avg. BIC: 4.3862 AIC : 19967.3437 Avg. AIC: 4.3750 ******************************************************


346

Note that the parameter values of the last BFGS iteration are different that those reported in the final results. This reflects two things - first, the data were scaled before doing the BFGS minimization, but the mle_results script takes this into account and reports the results using the original scaling. But also, the parameterization α = exp(α∗ ) is used to enforce the restriction that α > 0. The unrestricted parameter α∗ = log α is used to define the log-likelihood function, since the BFGS minimization algorithm does not do contrained minimization. To get the standard error and t-statistic of the estimate of α, we need to use the delta method. This is done inside mle_results, making use of the function parameterize.m . Likewise, here are NB-II results:

MPITB extensions found OBDV ====================================================== BFGSMIN final results Used analytic gradient -----------------------------------------------------STRONG CONVERGENCE Function conv 1 Param conv 1 Gradient conv 1 -----------------------------------------------------Objective function value 2.18496 Stepsize 0.0104394 13 iterations -----------------------------------------------------param 1.0375 0.3673 0.2136 0.2816 0.3027 0.0843

gradient change 0.0000 -0.0000 -0.0000 0.0000 0.0000 -0.0000 0.0000 -0.0000 0.0000 0.0000 -0.0000 0.0000


-0.0048 0.4780

0.0000 -0.0000

347

-0.0000 0.0000

****************************************************** Negative Binomial model, MEPS 1996 full data set MLE Estimation Results BFGS convergence: Normal convergence Average Log-L: -2.184962 Observations: 4564 constant pub. ins. priv. ins. sex age edu inc alpha

estimate -1.068 1.101 0.476 0.564 0.025 0.029 -0.000 1.613

st. err 0.161 0.095 0.081 0.050 0.002 0.009 0.000 0.055

t-stat -6.622 11.611 5.880 11.166 12.240 3.106 -0.176 29.099

p-value 0.000 0.000 0.000 0.000 0.000 0.002 0.861 0.000

Information Criteria CAIC : 20019.7439 Avg. CAIC: 4.3864 BIC : 20011.7439 Avg. BIC: 4.3847 AIC : 19960.3362 Avg. AIC: 4.3734 ****************************************************** • For the OBDV usage measurel, the NB-II model does a slightly better job than the NB-I model, in terms of the average log-likelihood and the information criteria (more on this last in a moment). • Note that both versions of the NB model fit much better than does the Poisson model (see 13.4.2). • The estimated α is highly significant. To check the plausibility of the NB-II model, we can compare the sample unconditional variance with the estimated unconditional variance according to the NB-II


model: Vd (y) =

n λ ˆ t +αˆ (λ ˆ t )2 ∑t=1 . n

348

For OBDV and ERV (estimation results not reported),

we get For OBDV, the overdispersion problem is significantly better than in the PoisTABLE 2. Marginal Variances, Sample and Estimated (NB-II) OBDV ERV Sample 38.09 0.151 Estimated 30.58 0.182

son case, but there is still some that is not captured. For ERV, the negative binomial model seems to capture the overdispersion adequately. 16.2.2. Finite mixture models: the mixed negative binomial model. The finite mixture approach to fitting health care demand was introduced by Deb and Trivedi (1997). The mixture approach has the intuitive appeal of allowing for subgroups of the population with different health status. If individuals are classified as healthy or unhealthy then two subgroups are defined. A finer classification scheme would lead to more subgroups. Many studies have incorporated objective and/or subjective indicators of health status in an effort to capture this heterogeneity. The available objective measures, such as limitations on activity, are not necessarily very informative about a person’s overall health status. Subjective, self-reported measures may suffer from the same problem, and may also not be exogenous Finite mixture models are conceptually simple. The density is fY (y, φ1 , ..., φ p, π1 , ..., π p−1) = p−1

p−1

∑ πi fY

i=1

(i)

p

(y, φi ) + π p fY (y, φ p ), p

where πi > 0, i = 1, 2, ..., p, π p = 1 − ∑i=1 πi , and ∑i=1 πi = 1. Identification requires

that the πi are ordered in some way, for example, π1 ≥ π2 ≥ · · · ≥ π p and φi 6= φ j , i 6= j.

This is simple to accomplish post-estimation by rearrangement and possible elimination of redundant component densities.


349

• The properties of the mixture density follow in a straightforward way from those of the components. In particular, the moment generating function is the same mixture of the moment generating functions of the component densities, p

so, for example, E(Y |x) = ∑i=1 πi µi (x), where µi (x) is the mean of the ith component density. • Mixture densities may suffer from overparameterization, since the total number of parameters grows rapidly with the number of component densities. It is possible to constrained parameters across the mixtures. • Testing for the number of component densities is a tricky issue. For example, testing for p = 1 (a single component, which is to say, no mixture) versus p = 2 (a mixture of two components) involves the restriction π1 = 1, which is on the boundary of the parameter space. Not that when π1 = 1, the parameters of the second component can take on any value without affecting the density. Usual methods such as the likelihood ratio test are not applicable when parameters are on the boundary under the null hypothesis. Information criteria means of choosing the model (see below) are valid. The following results are for a mixture of 2 NB-II models, for the OBDV data, which you can replicate using this program .

OBDV ****************************************************** Mixed Negative Binomial model, MEPS 1996 full data set MLE Estimation Results BFGS convergence: Normal convergence Average Log-L: -2.164783 Observations: 4564


constant pub. ins. priv. ins. sex age edu inc alpha constant pub. ins. priv. ins. sex age edu inc alpha Mix

estimate 0.127 0.861 0.146 0.346 0.024 0.025 -0.000 1.351 0.525 0.422 0.377 0.400 0.296 0.111 0.014 1.034 0.257

st. err 0.512 0.174 0.193 0.115 0.004 0.016 0.000 0.168 0.196 0.048 0.087 0.059 0.036 0.042 0.051 0.187 0.162

t-stat 0.247 4.962 0.755 3.017 6.117 1.590 -0.214 8.061 2.678 8.752 4.349 6.773 8.178 2.634 0.274 5.518 1.582

350

p-value 0.805 0.000 0.450 0.003 0.000 0.112 0.831 0.000 0.007 0.000 0.000 0.000 0.000 0.008 0.784 0.000 0.114

Information Criteria CAIC : 19920.3807 Avg. CAIC: 4.3647 BIC : 19903.3807 Avg. BIC: 4.3610 AIC : 19794.1395 Avg. AIC: 4.3370 ******************************************************

It is worth noting that the mixture parameter is not significantly different from zero, but also not that the coefficients of public insurance and age, for example, differ quite a bit between the two latent classes.

16.2.3. Information criteria. As seen above, a Poisson model can’t be tested (using standard methods) as a restriction of a negative binomial model. But it seems, based upon the values of the likelihood functions and the fact that the NB model fits the variance much better, that the NB model is more appropriate. How can we determine which of a set of competing models is the best?


351

The information criteria approach is one possibility. Information criteria are functions of the log-likelihood, with a penalty for the number of parameters used. Three popular information criteria are the Akaike (AIC), Bayes (BIC) and consistent Akaike (CAIC). The formulae are ˆ + k(ln n + 1) CAIC = −2 ln L(θ) ˆ + k ln n BIC = −2 ln L(θ) ˆ + 2k AIC = −2 ln L(θ) It can be shown that the CAIC and BIC will select the correctly specified model from a group of models, asymptotically. This doesn’t mean, of course, that the correct model is necesarily in the group. The AIC is not consistent, and will asymptotically favor an over-parameterized model over the correctly specified model. Here are information criteria values for the models we’ve seen, for OBDV. Pretty clearly, the NB models TABLE 3. Information Criteria, OBDV Model Poisson NB-I NB-II MNB-II

AIC 7.345 4.375 4.373 4.337

BIC 7.355 4.386 4.385 4.361

CAIC 7.357 4.388 4.386 4.365

are better than the Poisson. The one additional parameter gives a very significant improvement in the likelihood function value. Between the NB-I and NB-II models, the NB-II is slightly favored. But one should remember that information criteria values are statistics, with variances. With another sample, it may well be that the NB-I model would be favored, since the differences are so small. The MNB-II model is favored over the others, by all 3 information criteria.


352

Why is all of this in the chapter on QML? Let’s suppose that the correct model for OBDV is in fact the NB-II model. It turns out in this case that the Poisson model will give consistent estimates of the slope parameters (if a model is a member of the linearexponential family and the conditional mean is correctly specified, then the parameters of the conditional mean will be consistently estimated). So the Poisson estimator would be a QML estimator that is consistent for some parameters of the true model. The ordinary OPG or inverse Hessinan ”ML” covariance estimators are however biased and inconsistent, since the information matrix equality does not hold for QML estimators. But for i.i.d. data (which is the case for the MEPS data) the QML asymptotic covariance can be consistently estimated, as discussed above, using the sandwich form for the ML estimator. mle_results in fact reports sandwich results, so the Poisson estimation results would be reliable for inference even if the true model is the NB-I or NB-II. Not that they are in fact similar to the results for the NB models. However, if we assume that the correct model is the MNB-II model, as is favored by the information criteria, then both the Poisson and NB-x models will have misspecified mean functions, so the parameters that influence the means would be estimated with bias and inconsistently.

EXERCISES

353

Exercises Exercises (1) Considering the MEPS data (the description is in Section 13.4.2), for the OBDV (y) measure, let η be a latent index of health status that has expectation equal to unity.1 We suspect that η and PRIV may be correlated, but we assume that η is uncorrelated with the other regressors. We assume that E(y|PU B, PRIV, AGE, EDUC, INC, η) = exp(β1 + β2 PU B + β3 PRIV + β4 AGE + β5 EDUC + β6 INC)η. We use the Poisson QML estimator of the model y ∼ Poisson(λ) (16.2.2)

λ = exp(β1 + β2 PU B + β3 PRIV + β4 AGE + β5 EDUC + β6 INC).

Since much previous evidence indicates that health care services usage is overdispersed2, this is almost certainly not an ML estimator, and thus is not efficient. However, when η and PRIV are uncorrelated, this estimator is consistent for the β i parameters, since the conditional mean is correctly specified in that case. When η and PRIV are correlated, Mullahy’s (1997) NLIV estimator that uses the residual function ε=

y − 1, λ

1A restriction of this sort is necessary for identification.

2Overdispersion exists when the conditional variance is greater than the conditional mean. If this is the

case, the Poisson specification is not correct.

EXERCISES

354

where λ is defined in equation 16.2.2, with appropriate instruments, is consistent. As instruments we use all the exogenous regressors, as well as the cross products of PU B with the variables in Z = {AGE, EDUC, INC}. That is, the full set of instruments is W = {1 PU B Z PU B × Z }. (a) Calculate the Poisson QML estimates. (b) Calculate the generalized IV estimates (do it using a GMM formulation - see the portfolio example for hints how to do this). (c) Calculate the Hausman test statistic to test the exogeneity of PRIV. (d) comment on the results

CHAPTER 17

Nonlinear least squares (NLS) Readings: Davidson and MacKinnon, Ch. 2∗ and 5∗ ; Gallant, Ch. 1

17.1. Introduction and definition Nonlinear least squares (NLS) is a means of estimating the parameter of the model yt = f (xt , θ0 ) + εt . • In general, εt will be heteroscedastic and autocorrelated, and possibly nonnormally distributed. However, dealing with this is exactly as in the case of linear models, so we’ll just treat the iid case here, εt ∼ iid(0, σ2 ) If we stack the observations vertically, defining y = (y1 , y2 , ..., yn)0 f = ( f (x1 , θ), f (x1 , θ), ..., f (x1, θ))0 and ε = (ε1 , ε2 , ..., εn )0 we can write the n observations as y = f(θ) + ε 355

17.1. INTRODUCTION AND DEFINITION

356

Using this notation, the NLS estimator can be defined as 1 1 θˆ ≡ arg min sn (θ) = [y − f(θ)]0 [y − f(θ)] = k y − f(θ) k2 Θ n n • The estimator minimizes the weighted sum of squared errors, which is the same as minimizing the Euclidean distance between y and f(θ). The objective function can be written as

sn (θ) =

1 0 y y − 2y0 f(θ) + f(θ)0 f(θ) , n

which gives the first order conditions

∂ ˆ 0 ˆ ∂ ˆ 0 f(θ) y + f(θ) f(θ) ≡ 0. − ∂θ ∂θ

Define the n × K matrix (17.1.1)

ˆ ˆ ≡ Dθ0 f(θ). F(θ)

ˆ Using this, the first order conditions can be written In shorthand, use Fˆ in place of F(θ). as ˆ ≡ 0, −Fˆ 0 y + Fˆ 0 f(θ) or (17.1.2)

ˆ ≡ 0. Fˆ 0 y − f(θ)

This bears a good deal of similarity to the f.o.c. for the linear model - the derivative of the prediction is orthogonal to the prediction error. If f(θ) = Xθ, then Fˆ is simply X, so the f.o.c. (with spherical errors) simplify to X0 y − X0 Xβ = 0,

17.2. IDENTIFICATION

357

the usual 0LS f.o.c. We can interpret this geometrically: INSERT drawings of geometrical depiction of OLS and NLS (see Davidson and MacKinnon, pgs. 8,13 and 46). • Note that the nonlinearity of the manifold leads to potential multiple local maxima, minima and saddlepoints: the objective function sn (θ) is not necessarily well-behaved and may be difficult to minimize.

17.2. Identification As before, identification can be considered conditional on the sample, and asymptotically. The condition for asymptotic identification is that sn (θ) tend to a limiting function s∞ (θ) such that s∞ (θ0 ) < s∞ (θ), ∀θ 6= θ0 . This will be the case if s∞ (θ0 ) is strictly convex at θ0 , which requires that D2θ s∞ (θ0 ) be positive definite. Consider the objective function: sn (θ) =

1 n ∑ [yt − f (xt , θ)]2 n t=1

2 1 n f (xt , θ0 ) + εt − ft (xt , θ) = ∑ n t=1 = −

2 1 n 1 n 0 f (θ ) − f (θ) + ∑ (εt )2 t ∑ t n t=1 n t=1

2 n ft (θ0 ) − ft (θ) εt ∑ n t=1

• As in example 14.3, which illustrated the consistency of extremum estimators using OLS, we conclude that the second term will converge to a constant which does not depend upon θ. • A LLN can be applied to the third term to conclude that it converges pointwise to 0, as long as f(θ) and ε are uncorrelated.

17.2. IDENTIFICATION

358

• Next, pointwise convergence needs to be stregnthened to uniform almost sure convergence. There are a number of possible assumptions one could use. Here, we’ll just assume it holds. • Turning to the first term, we’ll assume a pointwise law of large numbers applies, so 2 a.s. 1 n ft (θ0 ) − ft (θ) → ∑ n t=1

(17.2.1)

Z

2 f (z, θ0 ) − f (z, θ) dµ(z),

where µ(x) is the distribution function of x. In many cases, f (x, θ) will be bounded and continuous, for all θ ∈ Θ, so strengthening to uniform almost

sure convergence is immediate. For example if f (x, θ) = [1 + exp(−xθ)] −1 , f : ℜK → (0, 1) , a bounded range, and the function is continuous in θ. Given these results, it is clear that a minimizer is θ0 . When considering identification (asymptotic), the question is whether or not there may be some other minimizer. A local condition for identification is that ∂2 ∂2 s (θ) = ∞ ∂θ∂θ0 ∂θ∂θ0

Z

2 f (x, θ0 ) − f (x, θ) dµ(x)

be positive definite at θ0 . Evaluating this derivative, we obtain (after a little work)

∂2 ∂θ∂θ0

Z

Z 0 0 f (x, θ ) − f (x, θ) dµ(x) = 2 Dθ f (z, θ0 )0 Dθ0 f (z, θ0 ) dµ(z) 0 2

θ

the expectation of the outer product of the gradient of the regression function evaluated at θ0 . (Note: the uniform boundedness we have already assumed allows passing the derivative through the integral, by the dominated convergence theorem.) This matrix will be positive definite (wp1) as long as the gradient vector is of full rank (wp1). The tangent space to the regression manifold must span a K -dimensional space if we are


359

to consistently estimate a K -dimensional parameter vector. This is analogous to the requirement that there be no perfect colinearity in a linear model. This is a necessary condition for identification. Note that the LLN implies that the above expectation is equal to F0 F J∞ (θ ) = 2 lim E n 0

17.3. Consistency We simply assume that the conditions of Theorem 19 hold, so the estimator is consistent. Given that the strong stochastic equicontinuity conditions hold, as discussed above, and given the above identification conditions an a compact estimation space (the closure of the parameter space Θ), the consistency proof’s assumptions are satisfied.

17.4. Asymptotic normality As in the case of GMM, we also simply assume that the conditions for asymptotic normality as in Theorem 22 hold. The only remaining problem is to determine the form of the asymptotic variance-covariance matrix. Recall that the result of the asymptotic normality theorem is √

d n θˆ − θ0 → N 0, J∞ (θ0 )−1 I∞ (θ0 )J∞ (θ0 )−1 ,

where J∞ (θ0 ) is the almost sure limit of

∂2 ∂θ∂θ0 sn (θ)

evaluated at θ0 , and

√

I∞ (θ0 ) = limVar nDθ sn (θ0 )

The objective function is 1 n sn (θ) = ∑ [yt − f (xt , θ)]2 n t=1


So Dθ sn (θ) = −

360

2 n ∑ [yt − f (xt , θ)] Dθ f (xt , θ). n t=1

Evaluating at θ0 , Dθ sn (θ0 ) = −

2 n ∑ εt Dθ f (xt , θ0). n t=1

Note that the expectation of this is zero, since εt and xt are assumed to be uncorrelated. So to calculate the variance, we can simply calculate the second moment about zero. Also note that n

∑ εt Dθ f (xt , θ0)

=

t=1

∂ 0 0 f(θ ) ε ∂θ

= F0 ε With this we obtain √

I∞ (θ0 ) = limVar nDθ sn (θ0 )

4 0 ’ F εε F n2 F0 F = 4σ2 lim E n = limnE

We’ve already seen that J∞ (θ0 ) = 2 lim E

F0 F , n

where the expectation is with respect to the joint density of x and ε. Combining these expressions for J∞ (θ0 ) and I∞ (θ0 ), and the result of the asymptotic normality theorem, we get

! 0 F −1 √ F d n θˆ − θ0 → N 0, lim E σ2 . n

17.5. EXAMPLE: THE POISSON MODEL FOR COUNT DATA

361

We can consistently estimate the variance covariance matrix using

(17.4.1)

Fˆ 0 Fˆ n

−1

σˆ 2 ,

where Fˆ is defined as in equation 17.1.1 and ˆ 0 y − f(θ) ˆ y − f(θ) σˆ = , n 2

the obvious estimator. Note the close correspondence to the results for the linear model.

17.5. Example: The Poisson model for count data Suppose that yt conditional on xt is independently distributed Poisson. A Poisson random variable is a count data variable, which means it can take the values {0,1,2,...}. This sort of model has been used to study visits to doctors per year, number of patents registered by businesses per year, etc. The Poisson density is y

f (yt ) =

exp(−λt )λt t , yt ∈ {0, 1, 2, ...}. yt !

The mean of yt is λt , as is the variance. Note that λt must be positive. Suppose that the true mean is λt0 = exp(xt0 β0 ), which enforces the positivity of λt . Suppose we estimate β0 by nonlinear least squares: n 2 ˆβ = arg min sn (β) = 1 yt − exp(xt0 β) ∑ T t=1

17.6. THE GAUSS-NEWTON ALGORITHM

362

We can write 2 1 n sn (β) = exp(xt0 β0 + εt − exp(xt0 β) ∑ T t=1 =

2 1 n 2 1 n 1 n 0 0 0 ε + 2 εt exp(xt0 β0 − exp(xt0 β) exp(x β − exp(x β) + ∑ ∑ ∑ t t t T t=1 T t=1 T t=1

The last term has expectation zero since the assumption that E (yt |xt ) = exp(xt0 β0 ) implies that E (εt |xt ) = 0, which in turn implies that functions of xt are uncorrelated

with εt . Applying a strong LLN, and noting that the objective function is continuous on a compact parameter space, we get s∞ (β) = Ex exp(x0 β0 − exp(x0 β)

2

+ Ex exp(x0 β0 )

where the last term comes from the fact that the conditional variance of ε is the same as the variance of y. This function is clearly minimized at β = β0 , so the NLS estimator is consistent as long as identification holds.

√ E XERCISE 27. Determine the limiting distribution of n βˆ − β0 . This means ∂2 0 ), ∂sn (β) , and I (β0 ). Again, use a finding the the specific forms of ∂β∂β s (β), J (β 0 n ∂β

CLT as needed, no need to verify that it can be applied.

17.6. The Gauss-Newton algorithm Readings: Davidson and MacKinnon, Chapter 6, pgs. 201-207 ∗ . The Gauss-Newton optimization technique is specifically designed for nonlinear least squares. The idea is to linearize the nonlinear model, rather than the objective function. The model is y = f(θ0 ) + ε.


363

At some θ in the parameter space, not equal to θ0 , we have y = f(θ) + ν where ν is a combination of the fundamental error term ε and the error due to evaluating the regression function at θ rather than the true value θ0 . Take a first order Taylor’s series approximation around a point θ1 : y = f(θ1 ) + Dθ0 f θ1 θ − θ1 + ν + approximation error.

Define z ≡ y − f(θ1 ) and b ≡ (θ − θ1 ). Then the last equation can be written as z = F(θ1 )b + ω, where, as above, F(θ1 ) ≡ Dθ0 f(θ1 ) is the n × K matrix of derivatives of the regression function, evaluated at θ1 , and ω is ν plus approximation error from the truncated

Taylor’s series.

• Note that F is known, given θ1 . • Note that one could estimate b simply by performing OLS on the above equation. ˆ we calculate a new round estimate of θ0 as θ2 = bˆ + θ1 . With this, • Given b, take a new Taylor’s series expansion around θ2 and repeat the process. Stop

when bˆ = 0 (to within a specified tolerance).

To see why this might work, consider the above approximation, but evaluated at the NLS estimator: ˆ + F(θ) ˆ θ − θˆ + ω y = f(θ)


364

The OLS estimate of b ≡ θ − θˆ is bˆ = Fˆ 0 Fˆ This must be zero, since

−1

ˆ . Fˆ 0 y − f(θ)

ˆ ≡0 Fˆ 0 θˆ y − f(θ)

by definition of the NLS estimator (these are the normal equations as in equation ˆ updating would stop. 17.1.2, Since bˆ ≡ 0 when we evaluate at θ,

• The Gauss-Newton method doesn’t require second derivatives, as does the Newton-Raphson method, so it’s faster. • The varcov estimator, as in equation 17.4.1 is simple to calculate, since we have Fˆ as a by-product of the estimation process (i.e., it’s just the last round “regressor matrix”). In fact, a normal OLS program will give the NLS varcov estimator directly, since it’s just the OLS varcov estimator from the last iteration. • The method can suffer from convergence problems since F(θ) 0 F(θ), may be very nearly singular, even with an asymptotically identified model, especially ˆ Consider the example if θ is very far from θ. y = β1 + β2 xt β3 + εt When evaluated at β2 ≈ 0, β3 has virtually no effect on the NLS objective function, so F will have rank that is “essentially” 2, rather than 3. In this case, F0 F will be nearly singular, so (F0 F)−1 will be subject to large roundoff errors.

17.7. APPLICATION: LIMITED DEPENDENT VARIABLES AND SAMPLE SELECTION

365

17.7. Application: Limited dependent variables and sample selection Readings: Davidson and MacKinnon, Ch. 15∗ (a quick reading is sufficient), J. Heckman, “Sample Selection Bias as a Specification Error”, Econometrica, 1979 (This is a classic article, not required for reading, and which is a bit out-dated. Nevertheless it’s a good place to start if you encounter sample selection problems in your research). Sample selection is a common problem in applied research. The problem occurs when observations used in estimation are sampled non-randomly, according to some selection scheme.

17.7.1. Example: Labor Supply. Labor supply of a person is a positive number of hours per unit time supposing the offer wage is higher than the reservation wage, which is the wage at which the person prefers not to work. The model (very simple, with t subscripts suppressed): • Characteristics of individual: x

• Latent labor supply: s∗ = x0 β + ω

• Offer wage: wo = z0 γ + ν

• Reservation wage: wr = q0 δ + η Write the wage differential as w∗ =

z 0 γ + ν − q0 δ + η

≡ r0 θ + ε We have the set of equations

s∗ = x 0 β + ω w∗ = r0 θ + ε.


Assume that

 

ω ε





 ∼ N 

0 0

  ,

σ2

ρσ

ρσ

1

366



 .

We assume that the offer wage and the reservation wage, as well as the latent variable s∗ are unobservable. What is observed is w = 1 [w∗ > 0] s = ws∗ . In other words, we observe whether or not a person is working. If the person is working, we observe labor supply, which is equal to latent labor supply, s ∗ . Otherwise, s = 0 6= s∗ . Note that we are using a simplifying assumption that individuals can freely choose their weekly hours of work. Suppose we estimated the model s∗ = x0 β + residual using only observations for which s > 0. The problem is that these observations are those for which w∗ > 0, or equivalently, −ε < r0 θ and

E ω| − ε < r0 θ 6= 0,

since ε and ω are dependent. Furthermore, this expectation will in general depend on x since elements of x can enter in r. Because of these two facts, least squares estimation is biased and inconsistent. Consider more carefully E [ω| − ε < r0 θ] . Given the joint normality of ω and ε, we can write (see for example Spanos Statistical Foundations of Econometric Modelling,


367

pg. 122) ω = ρσε + η, where η has mean zero and is independent of ε. With this we can write s∗ = x0 β + ρσε + η. If we condition this equation on −ε < r0 θ we get s = x0 β + ρσE (ε| − ε < r0 θ) + η which may be written as s = x0 β + ρσE (ε|ε > −r0 θ) + η • A useful result is that for z ∼ N(0, 1) E(z|z > z∗ ) =

φ(z∗ ) , Φ(−z∗ )

where φ (·) and Φ (·) are the standard normal density and distribution function, respectively. The quantity on the RHS above is known as the inverse Mill’s ratio: IMR(z∗ ) =

φ(z∗ ) Φ(−z∗ )

With this we can write (making use of the fact that the standard normal density is symmetric about zero, so that φ(−a) = φ(a)):


368

φ (r0 θ) +η Φ (r0 θ)   i β φ(r0 θ)   + η. 0 Φ(r θ) ζ

s = x0 β + ρσ

(17.7.1) (17.7.2)

≡

h

x0

where ζ = ρσ. The error term η has conditional mean zero, and is uncorrelated with the regressors x0

φ(r0 θ) Φ(r0 θ)

. At this point, we can estimate the equation by NLS.

• Heckman showed how one can estimate this in a two step procedure where

first θ is estimated, then equation 17.7.2 is estimated by least squares using the estimated value of θ to form the regressors. This is inefficient and estimation of the covariance is a tricky issue. It is probably easier (and more efficient) just to do MLE.

• The model presented above depends strongly on joint normality. There exist many alternative models which weaken the maintained assumptions. It is possible to estimate consistently without distributional assumptions. See Ahn and Powell, Journal of Econometrics, 1994.

CHAPTER 18

Nonparametric inference 18.1. Possible pitfalls of parametric inference: estimation Readings: H. White (1980) “Using Least Squares to Approximate Unknown Regression Functions,” International Economic Review, pp. 149-70. In this section we consider a simple example, which illustrates both why nonparametric methods may in some cases be preferred to parametric methods. We suppose that data is generated by random sampling of (y, x), where y = f (x) +ε, x is uniformly distributed on (0, 2π), and ε is a classical error. Suppose that f (x) = 1 +

3x x 2 − 2π 2π

The problem of interest is to estimate the elasticity of f (x) with respect to x, throughout the range of x. In general, the functional form of f (x) is unknown. One idea is to take a Taylor’s series approximation to f (x) about some point x0 . Flexible functional forms such as the transcendental logarithmic (usually know as the translog) can be interpreted as second order Taylor’s series approximations. We’ll work with a first order approximation, for simplicity. Approximating about x0 : h(x) = f (x0 ) + Dx f (x0 ) (x − x0 ) If the approximation point is x0 = 0, we can write h(x) = a + bx 369

18.1. POSSIBLE PITFALLS OF PARAMETRIC INFERENCE: ESTIMATION

370

The coefficient a is the value of the function at x = 0, and the slope is the value of the derivative at x = 0. These are of course not known. One might try estimation by ordinary least squares. The objective function is n

s(a, b) = 1/n ∑ (yt − h(xt ))2 . t=1

The limiting objective function, following the argument we used to get equations 14.3.1 and 17.2.1 is s∞ (a, b) =

Z 2π 0

( f (x) − h(x))2 dx.

The theorem regarding the consistency of extremum estimators (Theorem 19) tells us that aˆ and bˆ will converge almost surely to the values that minimize the limiting objective function. Solving the first order conditions1 reveals that s∞ (a, b) obtains its ˆ minimum at a0 = 67 , b0 = π1 . The estimated approximating function h(x) therefore

tends almost surely to

h∞ (x) = 7/6 + x/π In Figure 18.1.1 we see the true function and the limit of the approximation to see the asymptotic bias as a function of x. (The approximating model is the straight line, the true model has curvature.) Note that the approximating model is in general inconsistent, even at the approximation point. This shows that “flexible functional forms” based upon Taylor’s series approximations do not in general lead to consistent estimation of functions. The approximating model seems to fit the true model fairly well, asymptotically. However, we are interested in the elasticity of the function. Recall that an elasticity is

1The following results were obtained using the command maxima -b fff.mac You can get the source

file at http://pareto.uab.es/mcreel/Econometrics/Include/Nonparametric/fff.mac.


371

F IGURE 18.1.1. True and simple approximating functions 3.5

Fun1 x/%PI+7/6

3

2.5

2

1.5

1

0

1

2

3

4

5

6

7

the marginal function divided by the average function: ε(x) = xφ0 (x)/φ(x) Good approximation of the elasticity over the range of x will require a good approximation of both f (x) and f 0 (x) over the range of x. The approximating elasticity is η(x) = xh0 (x)/h(x) In Figure 18.1.2 we see the true elasticity and the elasticity obtained from the limiting approximating model. The true elasticity is the line that has negative slope for large x. Visually we see that the elasticity is not approximated so well. Root mean squared error in the approximation of the elasticity is Z

2π 0

2

(ε(x) − η(x)) dx

1/2

= . 31546


372

F IGURE 18.1.2. True and approximating elasticities 0.7

Fun1 x/(%PI*(x/%PI+7/6))

0.6

0.5

0.4

0.3

0.2

0.1

0

0

1

2

3

4

5

6

7

Now suppose we use the leading terms of a trigonometric series as the approximating model. The reason for using a trigonometric series as an approximating model is motivated by the asymptotic properties of the Fourier flexible functional form (Gallant, 1981, 1982), which we will study in more detail below. Normally with this type of model the number of basis functions is an increasing function of the sample size. Here we hold the set of basis function fixed. We will consider the asymptotic behavior of a fixed model, which we interpret as an approximation to the estimator’s behavior in finite samples. Consider the set of basis functions:

Z(x) =

h

1 x cos(x) sin(x) cos(2x) sin(2x)

The approximating model is gK (x) = Z(x)α.

i

.


373

F IGURE 18.1.3. True function and more flexible approximation 3.5

Fun1 Fun2

3

2.5

2

1.5

1

0

1

2

3

4

5

6

7

Maintaining these basis functions as the sample size increases, we find that the limiting objective function is minimized at

7 1 1 1 a1 = , a2 = , a3 = − 2 , a4 = 0, a5 = − 2 , a6 = 0 . 6 π π 4π

Substituting these values into gK (x) we obtain the almost sure limit of the approximation (18.1.1)

1 g∞ (x) = 7/6 + x/π + (cos x) − 2 π

1 + (sin x) 0 + (cos 2x) − 2 4π

+ (sin 2x) 0

In Figure 18.1.3 we have the approximation and the true function: Clearly the truncated trigonometric series model offers a better approximation, asymptotically, than does the linear model. In Figure 18.1.4 we have the more flexible approximation’s elasticity and that of the true function: On average, the fit is better, though there is

18.2. POSSIBLE PITFALLS OF PARAMETRIC INFERENCE: HYPOTHESIS TESTING

374

F IGURE 18.1.4. True elasticity and more flexible approximation 0.7

Fun1 Fun2

0.6

0.5

0.4

0.3

0.2

0.1

0

0

1

2

3

4

5

6

7

some implausible wavyness in the estimate. Root mean squared error in the approximation of the elasticity is Z 2π 0

g0 (x)x ε(x) − ∞ g∞ (x)

2

dx

!1/2

= . 16213,

about half that of the RMSE when the first order approximation is used. If the trigonometric series contained infinite terms, this error measure would be driven to zero, as we shall see.

18.2. Possible pitfalls of parametric inference: hypothesis testing What do we mean by the term “nonparametric inference”? Simply, this means inferences that are possible without restricting the functions of interest to belong to a parametric family. • Consider means of testing for the hypothesis that consumers maximize utility.

A consequence of utility maximization is that the Slutsky matrix D2p h(p,U ),

18.2. POSSIBLE PITFALLS OF PARAMETRIC INFERENCE: HYPOTHESIS TESTING

375

where h(p,U ) are the a set of compensated demand functions, must be negative semi-definite. One approach to testing for utility maximization would estimate a set of normal demand functions x(p, m). • Estimation of these functions by normal parametric methods requires specification of the functional form of demand, for example x(p, m) = x(p, m, θ0 ) + ε, θ0 ∈ Θ0 , where x(p, m, θ0 ) is a function of known form and Θ0 is a finite dimensional parameter. ˆ to calculate (by solving the in• After estimation, we could use xˆ = x(p, m, θ)

b 2p h(p,U ). If we can statistically tegrability problem, which is non-trivial) D reject that the matrix is negative semi-definite, we might conclude that con-

sumers don’t maximize utility. • The problem with this is that the reason for rejection of the theoretical proposition may be that our choice of functional form is incorrect. In the introductory section we saw that functional form misspecification leads to inconsistent estimation of the function and its derivatives. • Testing using parametric models always means we are testing a compound hypothesis. The hypothesis that is tested is 1) the economic proposition we wish to test, and 2) the model is correctly specified. Failure of either 1) or 2) can lead to rejection. This is known as the “model-induced augmenting hypothesis.” • Varian’s WARP allows one to test for utility maximization without specifying the form of the demand functions. The only assumptions used in the test are those directly implied by theory, so rejection of the hypothesis calls into question the theory.

18.3. THE FOURIER FUNCTIONAL FORM

376

• Nonparametric inference allows direct testing of economic propositions, without the “model-induced augmenting hypothesis”.

18.3. The Fourier functional form Readings: Gallant, 1987, “Identification and consistency in semi-nonparametric regression,” in Advances in Econometrics, Fifth World Congress, V. 1, Truman Bewley, ed., Cambridge. • Suppose we have a multivariate model y = f (x) + ε, where f (x) is of unknown form and x is a P−dimensional vector. For simplicity, assume that ε is a classical error. Let us take the estimation of the vector of elasticities with typical element ξ xi =

xi ∂ f (x) , f (x) ∂xi f (x)

at an arbitrary point xi . The Fourier form, following Gallant (1982), but with a somewhat different parameterization, may be written as

(18.3.1) gK (x | θK ) = α + x β + 1/2x Cx + 0

0

A

J

∑∑

α=1 j=1

u jα cos( jk0α x) − v jα sin( jk0α x) .

where the K-dimensional parameter vector (18.3.2)

θK = {α, β0 , vec∗ (C)0, u11 , v11 , . . . , uJA , vJA }0 .


377

• We assume that the conditioning variables x have each been transformed to lie in an interval that is shorter than 2π. This is required to avoid periodic behavior of the approximation, which is desirable since economic functions aren’t periodic. For example, subtract sample means, divide by the maxima of the conditioning variables, and multiply by 2π − eps, where eps is some positive number less than 2π in value. • The kα are ”elementary multi-indices” which are simply P− vectors formed

of integers (negative, positive and zero). The kα , α = 1, 2, ..., A are required to be linearly independent, and we follow the convention that the first non-zero

element be positive. For example h

0 1 −1 0 1

i0

is a potential multi-index to be used, but h

i0

0 −1 −1 0 1

is not since its first nonzero element is negative. Nor is h

0 2 −2 0 2

i0

a multi-index we would use, since it is a scalar multiple of the original multiindex. • We parameterize the matrix C differently than does Gallant because it simplifies things in practice. The cost of this is that we are no longer able to test a quadratic specification using nested testing.


378

The vector of first partial derivatives is

(18.3.3) Dx gK (x | θK ) = β + Cx +

A

J

∑∑

α=1 j=1

−u jα sin( jk0α x) − v jα cos( jk0α x) jkα

and the matrix of second partial derivatives is

(18.3.4)

D2x gK (x|θK ) = C +

A

J

∑∑

α=1 j=1

−u jα cos( jk0α x) + v jα sin( jk0α x) j2 kα k0α

To define a compact notation for partial derivatives, let λ be an N-dimensional multi-index with no negative elements. Define | λ |∗ as the sum of the elements of λ.

If we have N arguments x of the (arbitrary) function h(x), use Dλ h(x) to indicate a

certain partial derivative: λ

D h(x) ≡

∂|λ|

∗

∂xλ1 1 ∂xλ2 2 · · · ∂xλNN

h(x)

When λ is the zero vector, Dλ h(x) ≡ h(x). Taking this definition and the last few

equations into account, we see that it is possible to define (1 × K) vector Z λ (x) so that Dλ gK (x|θK ) = zλ (x)0 θK .

(18.3.5)

• Both the approximating model and the derivatives of the approximating model are linear in the parameters. • For the approximating model to the function (not derivatives), write g K (x|θK ) = z0 θK for simplicity.

The following theorem can be used to prove the consistency of the Fourier form. T HEOREM 28. [Gallant and Nychka, 1987] Suppose that hˆ n is obtained by maximizing a sample objective function sn (h) over HKn where HK is a subset of some


379

function space H on which is defined a norm k h k. Consider the following conditions: (a) Compactness: The closure of H with respect to k h k is compact in the relative topology defined by k h k. (b) Denseness: ∪K HK , K = 1, 2, 3, ... is a dense subset of the closure of H with respect to k h k and HK ⊂ HK+1 . (c) Uniform convergence: There is a point h∗ in H and there is a function s∞ (h, h∗ ) that is continuous in h with respect to k h k such that lim sup | sn (h) − s∞ (h, h∗ ) |= 0

n→∞

H

almost surely. (d) Identification: Any point h in the closure of H with s∞ (h, h∗ ) ≥ s∞ (h∗ , h∗ ) must have k h − h∗ k= 0.

Under these conditions limn→∞ k h∗ − hˆ n k= 0 almost surely, provided that limn→∞ Kn =

∞ almost surely.

The modification of the original statement of the theorem that has been made is to set the parameter space Θ in Gallant and Nychka’s (1987) Theorem 0 to a single point and to state the theorem in terms of maximization rather than minimization. This theorem is very similar in form to Theorem 19. The main differences are: (1) A generic norm k h k is used in place of the Euclidean norm. This norm may be stronger than the Euclidean norm, so that convergence with respect to k h k implies convergence w.r.t the Euclidean norm. Typically we will want to make sure that the norm is strong enough to imply convergence of all functions of interest.


380

(2) The “estimation space” H is a function space. It plays the role of the parameter space Θ in our discussion of parametric estimators. There is no restriction to a parametric family, only a restriction to a space of functions that satisfy certain conditions. This formulation is much less restrictive than the restriction to a parametric family. (3) There is a denseness assumption that was not present in the other theorem. We will not prove this theorem (the proof is quite similar to the proof of theorem [19], see Gallant, 1987) but we will discuss its assumptions, in relation to the Fourier form as the approximating model. 18.3.1. Sobolev norm. Since all of the assumptions involve the norm k h k , we need to make explicit what norm we wish to use. We need a norm that guarantees that the errors in approximation of the functions we are interested in are accounted for. Since we are interested in first-order elasticities in the present case, we need close approximation of both the function f (x) and its first derivative f 0 (x), throughout the range of x. Let X be an open set that contains all values of x that we’re interested in. The Sobolev norm is appropriate in this case. It is defined, making use of our notation for partial derivatives, as: k h km,X = max sup Dλ h(x) |λ∗ |≤m X

To see whether or not the function f (x) is well approximated by an approximating model gK (x | θK ), we would evaluate k f (x) − gK (x | θK ) km,X . We see that this norm takes into account errors in approximating the function and partial derivatives up to order m. If we want to estimate first order elasticities, as is the


381

case in this example, the relevant m would be m = 1. Furthermore, since we examine the sup over X , convergence w.r.t. the Sobolev means uniform convergence, so that we obtain consistent estimates for all values of x. 18.3.2. Compactness. Verifying compactness with respect to this norm is quite technical and unenlightening. It is proven by Elbadawi, Gallant and Souza, Econometrica, 1983. The basic requirement is that if we need consistency w.r.t. k h k m,X , then the functions of interest must belong to a Sobolev space which takes into account derivatives of order m + 1. A Sobolev space is the set of functions Wm,X (D) = {h(x) :k h(x) km,X < D},

where D is a finite constant. In plain words, the functions must have bounded partial derivatives of one order higher than the derivatives we seek to estimate. 18.3.3. The estimation space and the estimation subspace. Since in our case we’re interested in consistent estimation of first-order elasticities, we’ll define the estimation space as follows: D EFINITION 29. [Estimation space] The estimation space H = W2,X (D). The estimation space is an open set, and we presume that h∗ ∈ H . So we are assuming that the function to be estimated has bounded second derivatives throughout X . With seminonparametric estimators, we don’t actually optimize over the estimation space. Rather, we optimize over a subspace, HKn , defined as: D EFINITION 30. [Estimation subspace] The estimation subspace H K is defined as HK = {gK (x|θK ) : gK (x|θK ) ∈ W2,Z (D), θK ∈ ℜK },


382

where gK (x, θK ) is the Fourier form approximation as defined in Equation 18.3.1. 18.3.4. Denseness. The important point here is that HK is a space of functions that is indexed by a finite dimensional parameter (θK has K elements, as in equation 18.3.2). With n observations, n > K, this parameter is estimable. Note that the true function h∗ is not necessarily an element of HK , so optimization over HK may not lead to a consistent estimator. In order for optimization over HK to be equivalent to optimization over H , at least asymptotically, we need that: (1) The dimension of the parameter vector, dim θKn → ∞ as n → ∞. This is achieved by making A and J in equation 18.3.1 increasing functions of n, the sample size. It is clear that K will have to grow more slowly than n. The second requirement is: (2) We need that the HK be dense subsets of H . The estimation subspace HK , defined above, is a subset of the closure of the estimation space, H . A set of subsets Aa of a set A is “dense” if the closure of the countable union of the subsets is equal to the closure of A : ∪∞ a=1 Aa = A Use a picture here. The rest of the discussion of denseness is provided just for completeness: there’s no need to study it in detail. To show that HK is a dense subset of H with respect to k h k1,X , it is useful to apply Theorem 1 of Gallant (1982), who in

turn cites Edmunds and Moscatelli (1977). We reproduce the theorem as presented by Gallant, with minor notational changes, for convenience of reference: T HEOREM 31. [Edmunds and Moscatelli, 1977] Let the real-valued function h ∗ (x) be continuously differentiable up to order m on an open set containing the closure of X . Then it is possible to choose a triangular array of coefficients θ 1 , θ2 , . . . θK , . . .,


383

such that for every q with 0 ≤ q < m, and every ε > 0, k h∗ (x) − hK (x|θK ) kq,X =

o(K −m+q+ε ) as K → ∞.

In the present application, q = 1, and m = 2. By definition of the estimation space, the elements of H are once continuously differentiable on X , which is open and contains the closure of X , so the theorem is applicable. Closely following Gallant and Nychka (1987), ∪∞ HK is the countable union of the HK . The implication of Theorem 31 is that there is a sequence of {hK } from ∪∞ HK such that lim k h∗ − hK k1,X = 0,

K→∞

for all h∗ ∈ H . Therefore, H ⊂ ∪ ∞ HK .

However, ∪∞ HK ⊂ H , so ∪∞ HK ⊂ H . Therefore H = ∪ ∞ HK ,

so ∪∞ HK is a dense subset of H , with respect to the norm k h k1,X .

18.3.5. Uniform convergence. We now turn to the limiting objective function. We estimate by OLS. The sample objective function stated in terms of maximization is

1 n sn (θK ) = − ∑ (yt − gK (xt | θK ))2 n t=1


384

With random sampling, as in the case of Equations 14.3.1 and 17.2.1, the limiting objective function is

(18.3.6)

s∞ (g, f ) = −

Z

X

( f (x) − g(x))2 dµx − σ2ε .

where the true function f (x) takes the place of the generic function h ∗ in the presentation of the theorem. Both g(x) and f (x) are elements of ∪∞ HK . The pointwise convergence of the objective function needs to be strengthened to uniform convergence. We will simply assume that this holds, since the way to verify this depends upon the specific application. We also have continuity of the objective function in g, with respect to the norm k h k1,X since lim

kg1 −g0 k1,X →0

=

lim

s∞ g 1 , f ) − s ∞ g 0 , f )

Z h

kg1 −g0 k1,X →0 X

g1 (x) − f (x)

2

− g0 (x) − f (x)

2 i

dµx.

By the dominated convergence theorem (which applies since the finite bound D used to define W2,Z (D) is dominated by an integrable function), the limit and the integral can be interchanged, so by inspection, the limit is zero.

18.3.6. Identification. The identification condition requires that for any point (g, f ) in H × H , s∞ (g, f ) ≥ s∞ ( f , f ) ⇒ k g − f k1,X = 0. This condition is clearly satisfied given that g and f are once continuously differentiable (by the assumption that defines the estimation space).

18.3.7. Review of concepts. For the example of estimation of first-order elasticities, the relevant concepts are:


385

• Estimation space H = W2,X (D): the function space in the closure of which the true function must lie. • Consistency norm k h k1,X . The closure of H is compact with respect to this norm. • Estimation subspace HK . The estimation subspace is the subset of H that is

representable by a Fourier form with parameter θK . These are dense subsets of H .

• Sample objective function sn (θK ), the negative of the sum of squares. By standard arguments this converges uniformly to the • Limiting objective function s∞ ( g, f ), which is continuous in g and has a global maximum in its first argument, over the closure of the infinite union of the estimation subpaces, at g = f . • As a result of this, first order elasticities xi ∂ f (x) f (x) ∂xi f (x) are consistently estimated for all x ∈ X . 18.3.8. Discussion. Consistency requires that the number of parameters used in the expansion increase with the sample size, tending to infinity. If parameters are added at a high rate, the bias tends relatively rapidly to zero. A basic problem is that a high rate of inclusion of additional parameters causes the variance to tend more slowly to zero. The issue of how to chose the rate at which parameters are added and which to add first is fairly complex. A problem is that the allowable rates for asymptotic normality to obtain (Andrews 1991; Gallant and Souza, 1991) are very strict. Supposing we stick to these rates, our approximating model is: gK (x|θK ) = z0 θK .

18.4. KERNEL REGRESSION ESTIMATORS

386

• Define ZK as the n × K matrix of regressors obtained by stacking observations. The LS estimator is θˆ K = Z0K ZK

+

Z0K y,

where (·)+ is the Moore-Penrose generalized inverse. – This is used since Z0K ZK may be singular, as would be the case for K(n) large enough when some dummy variables are included. • . The prediction, z0 θˆ K , of the unknown function f (x) is asymptotically normally distributed: √ where

d n z0 θˆ K − f (x) → N(0, AV ),

# " + 0 ZK ZK AV = lim E z0 zσˆ 2 . n→∞ n

Formally, this is exactly the same as if we were dealing with a parametric linear model. I emphasize, though, that this is only valid if K grows very slowly as n grows. If we can’t stick to acceptable rates, we should probably use some other method of approximating the small sample distribution. Bootstrapping is a possibility. We’ll discuss this in the section on simulation. 18.4. Kernel regression estimators Readings: Bierens, 1987, “Kernel estimators of regression functions,” in Advances in Econometrics, Fifth World Congress, V. 1, Truman Bewley, ed., Cambridge. An alternative method to the semi-nonparametric method is a fully nonparametric method of estimation. Kernel regression estimation is an example (others are splines, nearest neighbor, etc.). We’ll consider the Nadaraya-Watson kernel regression estimator in a simple case.


387

• Suppose we have an iid sample from the joint density f (x, y), where x is k -dimensional. The model is yt = g(xt ) + εt , where E(εt |xt ) = 0. • The conditional expectation of y given x is g(x). By definition of the conditional expectation, we have g(x) =

Z

y

f (x, y) dy h(x)

1 = h(x)

Z

y f (x, y)dy,

where h(x) is the marginal density of x : h(x) =

Z

f (x, y)dy.

• This suggests that we could estimate g(x) by estimating h(x) and y f (x, y)dy. R

18.4.1. Estimation of the denominator. A kernel estimator for h(x) has the form 1 n K [(x − xt ) /γn ] ˆ , h(x) = ∑ n t=1 γkn where n is the sample size and k is the dimension of x. • The function K(·) (the kernel) is absolutely integrable: Z

|K(x)|dx < ∞,


388

and K(·) integrates to 1 : Z

K(x)dx = 1.

In this respect, K(·) is like a density function, but we do not necessarily restrict K(·) to be nonnegative. • The window width parameter, γn is a sequence of positive numbers that satisfies lim γn = 0

n→∞

lim nγkn = ∞

n→∞

So, the window width must tend to zero, but not too quickly. ˆ • To show pointwise consistency of h(x) for h(x), first consider the expectation of the estimator (since the estimator is an average of iid terms we only need to consider the expectation of a representative term): Z −k ˆ E h(x) = γn K [(x − z) /γn ] h(z)dz.

Change variables as z∗ = (x − z)/γn , so z = x − γn z∗ and | dzdz∗0 | = γkn , we obtain Z ∗ ∗ k ∗ ˆ γ−k E h(x) = n K (z ) h(x − γn z )γn dz

=

Z

K (z∗ ) h(x − γn z∗ )dz∗ .


389

Now, asymptotically, ˆ lim E h(x) =

n→∞

lim

n→∞

=

Z

=

Z

Z

K (z∗ ) h(x − γn z∗ )dz∗

lim K (z∗ ) h(x − γn z∗ )dz∗

n→∞

K (z∗ ) h(x)dz∗

= h(x)

Z

K (z∗ ) dz∗

= h(x), since γn → 0 and K (z∗ ) dz∗ = 1 by assumption. (Note: that we can pass the R

limit through the integral is a result of the dominated convergence theorem.. For this to hold we need that h(·) be dominated by an absolutely integrable function. ˆ • Next, considering the variance of h(x), we have, due to the iid assumption nγknV

1 n ˆ h(x) = nγkn 2 ∑ V n t=1 = γ−k n

K [(x − xt ) /γn ] γkn

1 n ∑ V {K [(x − xt ) /γn]} n t=1

• By the representative term argument, this is ˆ nγknV h(x) = γ−k n V {K [(x − z) /γn ]}

• Also, since V (x) = E(x2 ) − E(x)2 we have

n o 2 2 ˆ nγknV h(x) = γ−k E (K [(x − z) /γ ]) − γ−k n n n {E (K [(x − z) /γn ])} Z 2 Z 2 −k k −k = γn K [(x − z) /γn ] h(z)dz − γn γn K [(x − z) /γn ] h(z)dz =

Z

h i2 2 k b γ−k K [(x − z) /γ ] h(z)dz − γ E h(x) n n n


390

The second term converges to zero: γkn E

h i2 b h(x) → 0,

by the previous result regarding the expectation and the fact that γ n → 0. Therefore, lim nγk V n→∞ n

Z 2 ˆh(x) = lim γ−k n K [(x − z) /γn ] h(z)dz. n→∞

Using exactly the same change of variables as before, this can be shown to be lim nγk V n→∞ n Since both

R

Z ˆh(x) = h(x) [K(z∗ )]2 dz∗ .

[K(z∗ )]2 dz∗ and h(x) are bounded, this is bounded, and since

nγkn → ∞ by assumption, we have that ˆ V h(x) → 0.

• Since the bias and the variance both go to zero, we have pointwise consistency (convergence in quadratic mean implies convergence in probability).

18.4.2. Estimation of the numerator. To estimate

R

y f (x, y)dy, we need an esti-

mator of f (x, y). The estimator has the same form as the estimator for h(x), only with one dimension more: 1 n K∗ [(y − yt ) /γn , (x − xt ) /γn ] fˆ(x, y) = ∑ n t=1 γk+1 n The kernel K∗ (·) is required to have mean zero: Z

yK∗ (y, x) dy = 0


391

and to marginalize to the previous kernel for h(x) : Z

K∗ (y, x) dy = K(x).

With this kernel, we have Z

y fˆ(y, x)dy =

1 n K [(x − xt ) /γn ] ∑ yt n t=1 γkn

by marginalization of the kernel, so we obtain 1 g(x) ˆ = ˆ h(x) =

=

Z

y fˆ(y, x)dy

K[(x−xt )/γn ] 1 n n ∑t=1 yt γkn K[(x−x 1 n t )/γn ] n ∑t=1 γkn n yt K [(x − xt ) /γn ] ∑t=1 . n ∑t=1 K [(x − xt ) /γn ]

This is the Nadaraya-Watson kernel regression estimator. 18.4.3. Discussion. • The kernel regression estimator for g(xt ) is a weighted average of the y j , j = 1, 2, ..., n, where higher weights are associated with points that are closer to xt . The weights sum to 1. • The window width parameter γn imposes smoothness. The estimator is increasingly flat as γn → ∞, since in this case each weight tends to 1/n.

• A large window width reduces the variance (strong imposition of flatness), but increases the bias. • A small window width reduces the bias, but makes very little use of information except points that are in a small neighborhood of xt . Since relatively little information is used, the variance is large when the window width is small.

18.5. KERNEL DENSITY ESTIMATION

392

• The standard normal density is a popular choice for K(.) and K∗ (y, x), though there are possibly better alternatives. 18.4.4. Choice of the window width: Cross-validation. The selection of an appropriate window width is important. One popular method is cross validation. This consists of splitting the sample into two parts (e.g., 50%-50%). The first part is the “in sample” data, which is used for estimation, and the second part is the “out of sample” data, used for evaluation of the fit though RMSE or some other criterion. The steps are: (1) Split the data. The out of sample data is yout and xout . (2) Choose a window width γ. (3) With the in sample data, fit yˆtout corresponding to each xtout . This fitted value is a function of the in sample data, as well as the evaluation point xtout , but it does not involve ytout . (4) Repeat for all out of sample points. (5) Calculate RMSE(γ) (6) Go to step 2, or to the next step if enough window widths have been tried. (7) Select the γ that minimizes RMSE(γ) (Verify that a minimum has been found, for example by plotting RMSE as a function of γ). (8) Re-estimate using the best γ and all of the data. This same principle can be used to choose A and J in a Fourier form model.

18.5. Kernel density estimation The previous discussion suggests that a kernel density estimator may easily be constructed. We have already seen how joint densities may be estimated. If were interested in a conditional density, for example of y conditional on x, then the kernel

18.6. SEMI-NONPARAMETRIC MAXIMUM LIKELIHOOD

393

estimate of the conditional density is simply fby|x = =

=

fˆ(x, y) ˆ h(x) 1 n K∗ [(y−yt )/γn ,(x−xt )/γn ] n ∑t=1 γnk+1 1 n K[(x−xt )/γn ] n ∑t=1 γkn n 1 ∑t=1 K∗ [(y − yt ) /γn , (x − xt ) /γn ] n K [(x − xt ) /γn ] γn ∑t=1

where we obtain the expressions for the joint and marginal densities from the section on kernel regression.

18.6. Semi-nonparametric maximum likelihood Readings: Gallant and Nychka, Econometrica, 1987. For a Fortran program to do this and a useful discussion in the user’s guide, see this link . See also Cameron and Johansson, Journal of Applied Econometrics, V. 12, 1997. MLE is the estimation method of choice when we are confident about specifying the density. Is is possible to obtain the benefits of MLE when we’re not so confident about the specification? In part, yes. Suppose we’re interested in the density of y conditional on x (both may be vectors). Suppose that the density f (y|x, φ) is a reasonable starting approximation to the true density. This density can be reshaped by multiplying it by a squared polynomial. The new density is g p (y|x, φ, γ) =

h2p (y|γ) f (y|x, φ) η p (x, φ, γ)

where

p

h p (y|γ) =

∑ γk yk

k=0


394

and η p (x, φ, γ) is a normalizing factor to make the density integrate (sum) to one. Because h2p (y|γ)/η p (x, φ, γ) is a homogenous function of θ it is necessary to impose a normalization: γ0 is set to 1. The normalization factor η p (φ, γ) is calculated (following Cameron and Johansson) using E(Y r ) =

∞

∑ yr fY (y|φ, γ)

y=0 ∞

=

∑ yr

y=0 ∞

=

[h p (y|γ)]2 fY (y|φ) η p (φ, γ)

p

p

∑ ∑ ∑ yr fY (y|φ)γk γl yk yl /η p(φ, γ)

y=0 k=0 l=0 p

=

p

∞

∑ ∑ γk γl ∑ yr+k+l fY (y|φ)

k=0 l=0 p

=

(

y=0

)

/η p (φ, γ)

p

∑ ∑ γk γl mk+l+r /η p(φ, γ).

k=0 l=0

By setting r = 0 we get that the normalizing factor is 18.6.1 (18.6.1)

η p (φ, γ) =

p

p

∑ ∑ γk γl mk+l

k=0 l=0

Recall that γ0 is set to 1 to achieve identification. The mr in equation 18.6.1 are the raw moments of the baseline density. Gallant and Nychka (1987) give conditions under which such a density may be treated as correctly specified, asymptotically. Basically, the order of the polynomial must increase as the sample size increases. However, there are technicalities. Similarly to Cameron and Johannson (1997), we may develop a negative binomial polynomial (NBP) density for count data. The negative binomial baseline density may


395

be written (see equation as Γ(y + ψ) fY (y|φ) = Γ(y + 1)Γ(ψ)

ψ ψ+λ

ψ

λ ψ+λ

y

where φ = {λ, ψ}, λ > 0 and ψ > 0. The usual means of incorporating conditioning

variables x is the parameterization λ = ex β . When ψ = λ/α we have the negative 0

binomial-I model (NB-I). When ψ = 1/α we have the negative binomial-II (NP-II) model. For the NB-I density, V (Y ) = λ + αλ. In the case of the NB-II model, we have V (Y ) = λ + αλ2 . For both forms, E(Y ) = λ. The reshaped density, with normalization to sum to one, is (18.6.2)

[h p (y|γ)]2 Γ(y + ψ) fY (y|φ, γ) = η p (φ, γ) Γ(y + 1)Γ(ψ)

ψ ψ+λ

ψ

λ ψ+λ

y

.

To get the normalization factor, we need the moment generating function: (18.6.3)

MY (t) = ψψ λ − et λ + ψ

−ψ

.

To illustrate, Figure 18.6.1 shows calculation of the first four raw moments of the NB density, calculated using MuPAD, which is a Computer Algebra System that (use to be?) free for personal use. These are the moments you would need to use a second order polynomial (p = 2). MuPAD will output these results in the form of C code, which is relatively easy to edit to write the likelihood function for the model. This has been done in NegBinSNP.cc, which is a C++ version of this model that can be compiled to use with octave using the mkoctfile command. Note the impressive length of the expressions when the degree of the expansion is 4 or 5! This is an example of a model that would be difficult to formulate without the help of a program like MuPAD.


396

F IGURE 18.6.1. Negative binomial raw moments

It is possible that there is conditional heterogeneity such that the appropriate reshaping should be more local. This can be accomodated by allowing the γ k parameters to depend upon the conditioning variables, for example using polynomials. Gallant and Nychka, Econometrica, 1987 prove that this sort of density can approximate a wide variety of densities arbitrarily well as the degree of the polynomial increases with the sample size. This approach is not without its drawbacks: the sample objective function can have an extremely large number of local maxima that can lead to numeric difficulties. If someone could figure out how to do in a way such that the sample objective function was nice and smooth, they would probably get the paper published in a good journal. Any ideas? Here’s a plot of true and the limiting SNP approximations (with the order of the polynomial fixed) to four different count data densities, which variously exhibit over

18.7. EXAMPLES

397

and underdispersion, as well as excess zeros. The baseline model is a negative binomial density.

Case 1

Case 2

.5 .4

.1

.3 .2

.05

.1 0

5

Case 3

10

15

20

0

.25

.2

.2

.15

.15

Case 4

5

10

15

20

25

.1

.1

.05

.05 1

2

3

4

5

6

7

2.5

5

7.5

10

12.5

15

18.7. Examples We’ll use the MEPS OBDV data to illustrate kernel regression and semi-nonparametric maximum likelihood. 18.7.1. Kernel regression estimation. Let’s try a kernel regression fit for the OBDV data. The program OBDVkernel.m loads the MEPS OBDV data, scans over a range of window widths and calculates leave-one-out CV scores, and plots the fitted OBDV usage versus AGE, using the best window width. The plot is in Figure 18.7.1.

18.7. EXAMPLES

398

F IGURE 18.7.1. Kernel fitted OBDV usage versus AGE Kernel fit, OBDV visits versus AGE 5

4.5

4

3.5

3

2.5

2

15

20

25

30

35

40

45

50

55

60

65

Note that usage increases with age, just as we’ve seen with the parametric models. Once could use bootstrapping to generate a confidence interval to the fit. 18.7.2. Seminonparametric ML estimation and the MEPS data. Now let’s estimate a seminonparametric density for the OBDV data. We’ll reshape a negative binomial density, as discussed above. The program EstimateNBSNP.m loads the MEPS OBDV data and estimates the model, using a NB-I baseline density and a 2nd order polynomial expansion. The output is:

OBDV ====================================================== BFGSMIN final results Used numeric gradient

18.7. EXAMPLES

399

-----------------------------------------------------STRONG CONVERGENCE Function conv 1 Param conv 1 Gradient conv 1 -----------------------------------------------------Objective function value 2.17061 Stepsize 0.0065 24 iterations -----------------------------------------------------param 1.3826 0.2317 0.1839 0.2214 0.1898 0.0722 -0.0002 1.7853 -0.4358 0.1129

gradient 0.0000 -0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 -0.0000 0.0000 0.0000

change -0.0000 0.0000 0.0000 -0.0000 -0.0000 -0.0000 -0.0000 -0.0000 -0.0000 0.0000

****************************************************** NegBin SNP model, MEPS full data set MLE Estimation Results BFGS convergence: Normal convergence Average Log-L: -2.170614 Observations: 4564 constant pub. ins. priv. ins. sex age edu inc gam1 gam2 lnalpha

estimate -0.147 0.695 0.409 0.443 0.016 0.025 -0.000 1.785 -0.436 0.113

st. err 0.126 0.050 0.046 0.034 0.001 0.006 0.000 0.141 0.029 0.027

t-stat -1.173 13.936 8.833 13.148 11.880 3.903 -0.011 12.629 -14.786 4.166

p-value 0.241 0.000 0.000 0.000 0.000 0.000 0.991 0.000 0.000 0.000

18.7. EXAMPLES

400

Information Criteria CAIC : 19907.6244 Avg. CAIC: 4.3619 BIC : 19897.6244 Avg. BIC: 4.3597 AIC : 19833.3649 Avg. AIC: 4.3456 ****************************************************** Note that the CAIC and BIC are lower for this model than for the models presented in Table 3. This model fits well, still being parsimonious. You can play around trying other use measures, using a NP-II baseline density, and using other orders of expansions. Density functions formed in this way may have MANY local maxima, so you need to be careful before accepting the results of a casual run. To guard against having converged to a local maximum, one can try using multiple starting values, or one could try simulated annealing as an optimization method. If you uncomment the relevant lines in the program, you can use SA to do the minimization. This will take a lot of time, compared to the default BFGS minimization. The chapter on parallel computations might be interesting to read before trying this.

CHAPTER 19

Simulation-based estimation Readings: In addition to the book mentioned previously, articles include Gallant and Tauchen (1996), “Which Moments to Match?”, ECONOMETRIC THEORY, Vol. 12, 1996, pages 657-681;˘a Gourieroux, Monfort and Renault (1993), “Indirect Inference,” J. Apl. Econometrics; Pakes and Pollard (1989) Econometrica; McFadden (1989) Econometrica. 19.1. Motivation Simulation methods are of interest when the DGP is fully characterized by a parameter vector, but the likelihood function is not calculable. If it were available, we would simply estimate by MLE, which is asymptotically fully efficient. 19.1.1. Example: Multinomial and/or dynamic discrete response models. Let y∗i be a latent random vector of dimension m. Suppose that y∗i = Xi β + εi where Xi is m × K. Suppose that (19.1.1)

εi ∼ N(0, Ω)

Henceforth drop the i subscript when it is not needed for clarity. • y∗ is not observed. Rather, we observe a many-to-one mapping y = τ(y∗ ) 401

19.1. MOTIVATION

402

This mapping is such that each element of y is either zero or one (in some cases only one element will be one). • Define

Ai = A(yi ) = {y∗ |yi = τ(y∗ )}

Suppose random sampling of (yi , Xi ). In this case the elements of yi may not be independent of one another (and clearly are not if Ω is not diagonal). However, yi is independent of y j , i 6= j.

• Let θ = (β0 , (vec∗ Ω)0 )0 be the vector of parameters of the model. The contribution of the ith observation to the likelihood function is pi (θ) =

Z

Ai

n(y∗i − Xi β, Ω)dy∗i

where n(ε, Ω) = (2π)

−M/2

|Ω|

−1/2

−ε0 Ω−1 ε exp 2

is the multivariate normal density of an M -dimensional random vector. The log-likelihood function is 1 n ln L (θ) = ∑ ln pi (θ) n i=1 and the MLE θˆ solves the score equations ˆ 1 n Dθ pi (θ) 1 n ˆ g ( θ) = ≡ 0. i ∑ ∑ ˆ n i=1 n i=1 pi (θ) • The problem is that evaluation of Li (θ) and its derivative w.r.t. θ by standard methods of numeric integration such as quadrature is computationally infeasible when m (the dimension of y) is higher than 3 or 4 (as long as there are no restrictions on Ω).

19.1. MOTIVATION

403

• The mapping τ(y∗ ) has not been made specific so far. This setup is quite general: for different choices of τ(y∗ ) it nests the case of dynamic binary

discrete choice models as well as the case of multinomial discrete choice (the choice of one out of a finite set of alternatives). – Multinomial discrete choice is illustrated by a (very simple) job search model. We have cross sectional data on individuals’ matching to a set of m jobs that are available (one of which is unemployment). The utility of alternative j is u j = X jβ + ε j Utilities of jobs, stacked in the vector ui are not observed. Rather, we observe the vector formed of elements y j = 1 u j > uk , ∀k ∈ m, k 6= j

Only one of these elements is different than zero. – Dynamic discrete choice is illustrated by repeated choices over time between two alternatives. Let alternative j have utility u jt = W jt β − ε jt , j ∈ {1, 2} t ∈ {1, 2, ..., m} Then y∗ = u 2 − u 1 = (W2 −W1 )β + ε2 − ε1 ≡ Xβ + ε

19.1. MOTIVATION

404

Now the mapping is (element-by-element) y = 1 [y∗ > 0] , that is yit = 1 if individual i chooses the second alternative in period t, zero otherwise.

19.1.2. Example: Marginalization of latent variables. Economic data often presents substantial heterogeneity that may be difficult to model. A possibility is to introduce latent random variables. This can cause the problem that there may be no known closed form for the distribution of observable variables after marginalizing out the unobservable latent variables. For example, count data (that takes values 0, 1, 2, 3, ...) is often modeled using the Poisson distribution Pr(y = i) =

exp(−λ)λi i!

The mean and variance of the Poisson distribution are both equal to λ : E (y) = V (y) = λ.

Often, one parameterizes the conditional mean as λi = exp(Xi β). This ensures that the mean is positive (as it must be). Estimation by ML is straightforward. Often, count data exhibits “overdispersion” which simply means that V (y) > E (y).

19.1. MOTIVATION

405

If this is the case, a solution is to use the negative binomial distribution rather than the Poisson. An alternative is to introduce a latent variable that reflects heterogeneity into the specification: λi = exp(Xi β + ηi ) where ηi has some specified density with support S (this density may depend on additional parameters). Let dµ(ηi ) be the density of ηi . In some cases, the marginal density of y

exp [− exp(Xiβ + ηi )] [exp(Xiβ + ηi )]yi Pr(y = yi ) = dµ(ηi ) yi ! S Z

will have a closed-form solution (one can derive the negative binomial distribution in the way if η has an exponential distribution), but often this will not be possible. In this case, simulation is a means of calculating Pr(y = i), which is then used to do ML estimation. This would be an example of the Simulated Maximum Likelihood (SML) estimation. • In this case, since there is only one latent variable, quadrature is probably a better choice. However, a more flexible model with heterogeneity would allow all parameters (not just the constant) to vary. For example exp [− exp(Xi βi )] [exp(Xi βi )]yi Pr(y = yi ) = dµ(βi ) yi ! S Z

entails a K = dim βi -dimensional integral, which will not be evaluable by quadrature when K gets large.

19.1.3. Estimation of models specified in terms of stochastic differential equations. It is often convenient to formulate models in terms of continuous time using differential equations. A realistic model should account for exogenous shocks to the system, which can be done by assuming a random component. This leads to a model

19.1. MOTIVATION

406

that is expressed as a system of stochastic differential equations. Consider the process dyt = g(θ, yt )dt + h(θ, yt )dWt which is assumed to be stationary. {Wt } is a standard Brownian motion (Weiner process), such that W (T ) =

Z T 0

dWt ∼ N(0, T )

Brownian motion is a continuous-time stochastic process such that • W (0) = 0 • [W (s) −W (t)] ∼ N(0, s − t) • [W (s) −W (t)] and [W ( j) −W (k)] are independent for s > t > j > k. That is, non-overlapping segments are independent. One can think of Brownian motion the accumulation of independent normally distributed shocks with infinitesimal variance. • The function g(θ, yt ) is the deterministic part. • h(θ, yt ) determines the variance of the shocks. To estimate a model of this sort, we typically have data that are assumed to be observations of yt in discrete points y1 , y2 , ...yT . That is, though yt is a continuous process it is observed in discrete time. To perform inference on θ, direct ML or GMM estimation is not usually feasible, because one cannot, in general, deduce the transition density f (yt |yt−1 , θ). This density is necessary to evaluate the likelihood function or to evaluate moment conditions (which are based upon expectations with respect to this density). • A typical solution is to “discretize” the model, by which we mean to find a discrete time approximation to the model. The discretized version of the

19.2. SIMULATED MAXIMUM LIKELIHOOD (SML)

407

model is yt − yt−1 = g(φ, yt−1 ) + h(φ, yt−1 )εt εt ∼ N(0, 1) The discretization induces a new parameter, φ (that is, the φ0 which defines the best approximation of the discretization to the actual (unknown) discrete time version of the model is not equal to θ0 which is the true parameter value). This is an approximation, and as such “ML” estimation of φ (which is actually quasi-maximum likelihood, QML) based upon this equation is in general biased and inconsistent for the original parameter, θ. Nevertheless, the approximation shouldn’t be too bad, which will be useful, as we will see. • The important point about these three examples is that computational difficulties prevent direct application of ML, GMM, etc. Nevertheless the model is fully specified in probabilistic terms up to a parameter vector. This means that the model is simulable, conditional on the parameter vector.

19.2. Simulated maximum likelihood (SML) For simplicity, consider cross-sectional data. An ML estimator solves n ˆθML = arg max sn (θ) = 1 ∑ ln p(yt |Xt , θ) n t=1

where p(yt |Xt , θ) is the density function of the t th observation. When p(yt |Xt , θ) does not have a known closed form, θˆ ML is an infeasible estimator. However, it may be

possible to define a random function such that Eν f (ν, yt , Xt , θ) = p(yt |Xt , θ)


408

where the density of ν is known. If this is the case, the simulator 1 H p˜ (yt , Xt , θ) = ∑ f (νts , yt , Xt , θ) H s=1 is unbiased for p(yt |Xt , θ). • The SML simply substitutes p˜ (yt , Xt , θ) in place of p(yt |Xt , θ) in the loglikelihood function, that is 1 n θˆ SML = arg max sn (θ) = ∑ ln p˜ (yt , Xt , θ) n i=1 19.2.1. Example: multinomial probit. Recall that the utility of alternative j is u j = X jβ + ε j and the vector y is formed of elements y j = 1 u j > uk , k ∈ m, k 6= j

The problem is that Pr(y j = 1|θ) can’t be calculated when m is larger than 4 or 5. However, it is easy to simulate this probability. • Draw ε˜ i from the distribution N(0, Ω)

• Calculate u˜i = Xi β + ε˜ i (where Xi is the matrix formed by stacking the Xi j ) • Define y˜i j = 1 ui j > uik , ∀k ∈ m, k 6= j

• Repeat this H times and define

e πi j =

∑H h=1 y˜i jh H

• Define e πi as the m-vector formed of the e πi j . Each element of e πi is between 0 and 1, and the elements sum to one.

πi • Now p˜ (yi , Xi , θ) = y0i e


409

• The SML multinomial probit log-likelihood function is 1 n 0 ln L (β, Ω) = ∑ yi ln p˜ (yi , Xi , θ) n i=1 This is to be maximized w.r.t. β and Ω. Notes:

• The H draws of ε˜ i are draw only once and are used repeatedly during the

ˆ The draws are different for each i. If the ε˜ i are iterations used to find βˆ and Ω. re-drawn at every iteration the estimator will not converge.

• The log-likelihood function with this simulator is a discontinuous function of

β and Ω. This does not cause problems from a theoretical point of view since it can be shown that ln L (β, Ω) is stochastically equicontinuous. However, it does cause problems if one attempts to use a gradient-based optimization method such as Newton-Raphson.

• It may be the case, particularly if few simulations, H, are used, that some

elements of e πi are zero. If the corresponding element of yi is equal to 1, there will be a log(0) problem.

• Solutions to discontinuity: – 1) use an estimation method that doesn’t require a continuous and differentiable objective function, for example, simulated annealing. This is computationally costly. – 2) Smooth the simulated probabilities so that they are continuous functions of the parameters. For example, apply a kernel transformation such as

m m y˜i j = Φ A × ui j − max uik + .5 × 1 ui j = max uik k=1

k=1


410

where A is a large positive number. This approximates a step function such that y˜i j is very close to zero if ui j is not the maximum, and ui j = 1 if it is the maximum. This makes y˜i j a continuous function of β and Ω, so that p˜i j and therefore ln L (β, Ω) will be continuous and differentiable. p

Consistency requires that A(n) → ∞, so that the approximation to a step function becomes arbitrarily close as the sample size increases. There are alternative methods (e.g., Gibbs sampling) that may work better, but this is too technical to discuss here. • To solve to log(0) problem, one possibility is to search the web for the slog function. Also, increase H if this is a serious problem.

19.2.2. Properties. The properties of the SML estimator depend on how H is set. The following is taken from Lee (1995) “Asymptotic Bias in Simulated Maximum Likelihood Estimation of Discrete Choice Models,” Econometric Theory, 11, pp. 43783. T HEOREM 32. [Lee] 1) if limn→∞ n1/2 /H = 0, then d √ n θˆ SML − θ0 → N(0, I −1 (θ0 ))

2) if limn→∞ n1/2 /H = λ, λ a finite constant, then

d √ n θˆ SML − θ0 → N(B, I −1 (θ0 ))

where B is a finite vector of constants.

• This means that the SML estimator is asymptotically biased if H doesn’t grow faster than n1/2 .

19.3. METHOD OF SIMULATED MOMENTS (MSM)

411

• The varcov is the typical inverse of the information matrix, so that as long as H grows fast enough the estimator is consistent and fully asymptotically efficient.

19.3. Method of simulated moments (MSM) Suppose we have a DGP(y|x, θ) which is simulable given θ, but is such that the density of y is not calculable. Once could, in principle, base a GMM estimator upon the moment conditions mt (θ) = [K(yt , xt ) − k(xt , θ)] zt where k(xt , θ) =

Z

K(yt , xt )p(y|xt , θ)dy,

zt is a vector of instruments in the information set and p(y|xt , θ) is the density of y conditional on xt . The problem is that this density is not available. • However k(xt , θ) is readily simulated using 1 H e yth , xt ) k (xt , θ) = ∑ K(e H h=1

a.s. • By the law of large numbers, e k (xt , θ) → k (xt , θ) , as H → ∞, which provides

a clear intuitive basis for the estimator, though in fact we obtain consistency even for H finite, since a law of large numbers is also operating across the n observations of real data, so errors introduced by simulation cancel themselves out.

• This allows us to form the moment conditions (19.3.1)

h i m ft (θ) = K(yt , xt ) − e k (xt , θ) zt


412

where zt is drawn from the information set. As before, form 1 n m(θ) e = ft (θ) ∑m n i=1 " # 1 H 1 n = ∑ K(yt , xt ) − H ∑ k(eyth, xt ) zt n i=1 h=1

(19.3.2)

with which we form the GMM criterion and estimate as usual. Note that the unbiased simulator k(e yth , xt ) appears linearly within the sums. 19.3.1. Properties. Suppose that the optimal weighting matrix is used. McFadden (ref. above) and Pakes and Pollard (refs. above) show that the asymptotic distribution of the MSM estimator is very similar to that of the infeasible GMM estimator. In particular, assuming that the optimal weighting matrix is used, and for H finite, (19.3.3)

√

1 0 d −1 0 −1 ˆ n θMSM − θ → N 0, 1 + D∞ Ω D∞ H

where D∞ Ω−1 D0∞

−1

is the asymptotic variance of the infeasible GMM estimator.

• That is, the asymptotic variance is inflated by a factor 1 + 1/H. For this reason the MSM estimator is not fully asymptotically efficient relative to the infeasible GMM estimator, for H finite, but the efficiency loss is small and controllable, by setting H reasonably large. • The estimator is asymptotically unbiased even for H = 1. This is an advantage relative to SML. • If one doesn’t use the optimal weighting matrix, the asymptotic varcov is just the ordinary GMM varcov, inflated by 1 + 1/H. • The above presentation is in terms of a specific moment condition based upon the conditional mean. Simulated GMM can be applied to moment conditions of any form.


413

19.3.2. Comments. Why is SML inconsistent if H is finite, while MSM is? The reason is that SML is based upon an average of logarithms of an unbiased simulator (the densities of the observations). To use the multinomial probit model as an example, the log-likelihood function is ln L (β, Ω) =

1 n 0 ∑ yi ln pi(β, Ω) n i=1

ln L (β, Ω) =

1 n 0 ∑ yi ln p˜i(β, Ω) n i=1

The SML version is

The problem is that E ln( p˜i (β, Ω)) 6= ln(E p˜i (β, Ω)) in spite of the fact that E p˜i (β, Ω) = pi (β, Ω)

due to the fact that ln(·) is a nonlinear transformation. The only way for the two to be equal (in the limit) is if H tends to infinite so that p˜ (·) tends to p (·). The reason that MSM does not suffer from this problem is that in this case the unbiased simulator appears linearly within every sum of terms, and it appears within a sum over n (see equation [19.3.2]). Therefore the SLLN applies to cancel out simulation errors, from which we get consistency. That is, using simple notation for the random sampling case, the moment conditions " # 1 H 1 n (19.3.4) m(θ) ˜ = ∑ K(yt , xt ) − H ∑ k(eyth , xt ) zt n i=1 h=1 " # H 1 n 1 (19.3.5) = ∑ k(xt , θ0) + εt − H ∑ [k(xt , θ) + ε˜ ht ] zt n i=1 h=1

19.4. EFFICIENT METHOD OF MOMENTS (EMM)

414

converge almost surely to m˜ ∞ (θ) =

Z

k(x, θ0 ) − k(x, θ) z(x)dµ(x).

(note: zt is assume to be made up of functions of xt ). The objective function converges to s∞ (θ) = m˜ ∞ (θ)0 Ω−1 ˜ ∞ (θ) ∞ m which obviously has a minimum at θ0 , henceforth consistency. • If you look at equation 19.3.5 a bit, you will see why the variance inflation factor is (1 + H1 ).

19.4. Efficient method of moments (EMM) The choice of which moments upon which to base a GMM estimator can have very pronounced effects upon the efficiency of the estimator. • A poor choice of moment conditions may lead to very inefficient estimators, and can even cause identification problems (as we’ve seen with the GMM problem set). • The drawback of the above approach MSM is that the moment conditions used in estimation are selected arbitrarily. The asymptotic efficiency of the estimator may be low. • The asymptotically optimal choice of moments would be the score vector of the likelihood function, mt (θ) = Dθ ln pt (θ | It ) As before, this choice is unavailable.


415

The efficient method of moments (EMM) (see Gallant and Tauchen (1996), “Which Moments to Match?”, ECONOMETRIC THEORY, Vol. 12, 1996, pages 657-681) seeks to provide moment conditions that closely mimic the score vector. If the approximation is very good, the resulting estimator will be very nearly fully efficient. The DGP is characterized by random sampling from the density p(yt |xt , θ0 ) ≡ pt (θ0 ) We can define an auxiliary model, called the “score generator”, which simply provides a (misspecified) parametric density f (y|xt , λ) ≡ ft (λ) • This density is known up to a parameter λ. We assume that this density function is calculable. Therefore quasi-ML estimation is possible. Specifically, n ˆλ = arg max sn (λ) = 1 ∑ ln ft (λ). n t=1 Λ

ˆ ˆ we can calculate the score functions D ln f (yt |xt , λ). • After determining λ λ • The important point is that even if the density is misspecified, there is a

pseudo-true λ0 for which the true expectation, taken with respect to the true

but unknown density of y, p(y|xt , θ0 ), and then marginalized over x is zero: 0

0

∃λ : EX EY |X Dλ ln f (y|x, λ ) =

Z Z

X Y |X

Dλ ln f (y|x, λ0 )p(y|x, θ0 )dydµ(x) = 0

p 0 ˆ → • We have seen in the section on QML that λ λ ; this suggests using the

moment conditions (19.4.1)

1 n ˆ mn (θ, λ) = ∑ n t=1

Z

ˆ t (θ)dy Dλ ln ft (λ)p


416

• These moment conditions are not calculable, since pt (θ) is not available, but they are simulable using ˆ = m fn (θ, λ)

1 n 1 H ˆ Dλ ln f (e yth |xt , λ) ∑ ∑ n t=1 H h=1

where y˜th is a draw from DGP(θ), holding xt fixed. By the LLN and the fact ˆ converges to λ0 , that λ m e ∞ (θ0 , λ0 ) = 0.

This is not the case for other values of θ, assuming that λ0 is identified. • The advantage of this procedure is that if f (yt |xt , λ) closely approximates ˆ will closely approximate the optimal moment conp(y|xt , θ), then m e n (θ, λ)

ditions which characterize maximum likelihood estimation, which is fully efficient.

• If one has prior information that a certain density approximates the data well, it would be a good choice for f (·). • If one has no density in mind, there exist good ways of approximating unknown distributions parametrically: Philips’ ERA’s (Econometrica, 1983) and Gallant and Nychka’s (Econometrica, 1987) SNP density estimator which we saw before. Since the SNP density is consistent, the efficiency of the indirect estimator is the same as the infeasible ML estimator.

19.4.1. Optimal weighting matrix. I will present the theory for H finite, and possibly small. This is done because it is sometimes impractical to estimate with H very large. Gallant and Tauchen give the theory for the case of H so large that it may be treated as infinite (the difference being irrelevant given the numerical precision of


417

a computer). The theory for the case of H infinite follows directly from the results presented here. ˆ depends on the pseudo-ML estimate λ. ˆ We can The moment condition m(θ, e λ)

apply Theorem 22 to conclude that

√ ˆ d n λ − λ0 → N 0, J (λ0 )−1 I (λ0 )J (λ0 )−1

(19.4.2)

ˆ were in fact the true density p(y|xt , θ), then λ ˆ would be the If the density f (yt |xt , λ)

maximum likelihood estimator, and J (λ0 )−1 I (λ0 ) would be an identity matrix, due to the information matrix equality. However, in the present case we assume that

ˆ is only an approximation to p(y|xt , θ), so there is no cancellation. f (yt |xt , λ) 2 ∂ 0 ) . Comparing the definition of s (λ) with Recall that J (λ0 ) ≡ p lim ∂λ∂λ s (λ 0 n n

the definition of the moment condition in Equation 19.4.1, we see that J (λ0 ) = Dλ0 m(θ0 , λ0 ).

As in Theorem 22, ∂sn (λ) ∂sn (λ) I (λ ) = lim E n . n→∞ ∂λ λ0 ∂λ0 λ0 0

In this case, this is simply the asymptotic variance covariance matrix of the moment √ ˆ conditions, Ω. Now take a first order Taylor’s series approximation to nmn (θ0 , λ) about λ0 : √ 0 0 ˆ = √nm˜ n (θ0 , λ0 ) + √nD 0 m(θ ˆ − λ0 + o p (1) nm˜ n (θ0 , λ) ˜ , λ ) λ λ

First consider

√ nm˜ n (θ0 , λ0 ). It is straightforward but somewhat tedious to show

that the asymptotic variance of this term is

1 0 H I∞ (λ ).


Next consider the second term J (λ0 ), so we have

√

418

ˆ − λ0 . Note that D 0 m˜ n (θ0 , λ0 ) a.s. nDλ0 m(θ ˜ 0 , λ0 ) λ → λ

√ ˆ − λ0 , a.s. ˆ − λ0 = √nJ (λ0 ) λ nDλ0 m(θ ˜ 0 , λ0 ) λ But noting equation 19.4.2 √

0 a ˆ nJ (λ ) λ − λ ∼ N 0, I (λ0) 0

Now, combining the results for the first and second terms,

√ 1 0 ˆ a 0 nm˜ n (θ , λ) ∼ N 0, 1 + I (λ ) H Suppose that I[ (λ0 ) is a consistent estimator of the asymptotic variance-covariance matrix of the moment conditions. This may be complicated if the score generator is a poor approximator, since the individual score contributions may not have mean zero in this case (see the section on QML) . Even if this is the case, the individuals means can be calculated by simulation, so it is always possible to consistently estimate I (λ 0 ) when the model is simulable. On the other hand, if the score generator is taken to be correctly specified, the ordinary estimator of the information matrix is consistent. Combining this with the result on the efficient GMM weighting matrix in Theorem 25, we see that defining θˆ as ˆ 0 θˆ = arg min mn (θ, λ) Θ

1 1+ H

−1 [ ˆ 0 I (λ ) mn (θ, λ)

is the GMM estimator with the efficient choice of weighting matrix. • If one has used the Gallant-Nychka ML estimator as the auxiliary model, the appropriate weighting matrix is simply the information matrix of the auxiliary model, since the scores are uncorrelated. (e.g., it really is ML estimation


419

asymptotically, since the score generator can approximate the unknown density arbitrarily well). 19.4.2. Asymptotic distribution. Since we use the optimal weighting matrix, the asymptotic distribution is as in Equation 15.4.1, so we have (using the result in Equation 19.4.2): √ where

n θˆ − θ

0



d

→ N 0, D∞

1+

1 H

I (λ0 )

−1

!−1  , D0∞

D∞ = lim E Dθ m0n (θ0 , λ0 ) . n→∞

This can be consistently estimated using

ˆ ˆ λ) Dˆ = Dθ m0n (θ, 19.4.3. Diagnotic testing. The fact that √ 1 0 ˆ a 0 nmn (θ , λ) ∼ N 0, 1 + I (λ ) H implies that ˆ ˆ λ) nmn (θ,

0

1 1+ H

ˆ I (λ)

−1

a 2 ˆ ∼ ˆ λ) mn (θ, χ (q)

where q is dim(λ) − dim(θ), since without dim(θ) moment conditions the model is not identified, so testing is impossible. One test of the model is simply based on this statistic: if it exceeds the χ2 (q) critical point, something may be wrong (the small sample performance of this sort of test would be a topic worth investigating). • Information about what is wrong can be gotten from the pseudo-t-statistics: diag

1 1+ H

ˆ I (λ)

1/2 !−1

√

ˆ ˆ λ) nmn (θ,

19.5. EXAMPLES

420

can be used to test which moments are not well modeled. Since these moments are related to parameters of the score generator, which are usually related to certain features of the model, this information can be used to revise √ ˆ the model. These aren’t actually distributed as N(0, 1), since nmn (θ0 , λ) √ ˆ have different distributions (that of √nmn (θ, ˆ is somewhat ˆ λ) ˆ λ) and nmn (θ, more complicated). It can be shown that the pseudo-t statistics are biased toward nonrejection. See Gourieroux et. al. or Gallant and Long, 1995, for more details.

19.5. Examples 19.5.1. Estimation of stochastic differential equations. It is often convenient to formulate theoretical models in terms of differential equations, and when the observation frequency is high (e.g., weekly, daily, hourly or real-time) it may be more natural to adopt this framework for econometric models of time series. The most common approach to estimation of stochastic differential equations is to “discretize” the model, as above, and estimate using the discretized version. However, since the discretization is only an approximation to the true discrete-time version of the model (which is not calculable), the resulting estimator is in general biased and inconsistent. An alternative is to use indirect inference: The discretized model is used as the score generator. That is, one estimates by QML to obtain the scores of the discretized approximation:

yt − yt−1 = g(φ, yt−1 ) + h(φ, yt−1 )εt εt ∼ N(0, 1)

19.5. EXAMPLES

421

ˆ Then the system of stochastic differential equations Indicate these scores by mn (θ, φ). dyt = g(θ, yt )dt + h(θ, yt )dWt is simulated over θ, and the scores are calculated and averaged over the simulations N ˆ = 1 ∑ min (θ, φ) ˆ m˜ n (θ, φ) N i=1

θˆ is chosen to set the simulated scores to zero ˆ φ) ˆ ≡0 m˜ n (θ, (since θ and φ are of the same dimension). This method requires simulating the stochastic differential equation. There are many ways of doing this. Basically, they involve doing very fine discretizations: yt+τ = yt + g(θ, yt ) + h(θ, yt )ηt ηt ∼ N(0, τ) By setting τ very small, the sequence of ηt approximates a Brownian motion fairly well. This is only one method of using indirect inference for estimation of differential equations. There are others (see Gallant and Long, 1995 and Gourieroux et. al.). Use of a series approximation to the transitional density as in Gallant and Long is an interesting possibility since the score generator may have a higher dimensional parameter than the model, which allows for diagnostic testing. In the method described above the score generator’s parameter φ is of the same dimension as is θ, so diagnostic testing is not possible.

19.5. EXAMPLES

422

19.5.2. EMM estimation of a discrete choice model. In this section consider EMM estimation. There is a sophisticated package by Gallant and Tauchen for this, but here we’ll look at some simple, but hopefully didactic code. The file probitdgp.m generates data that follows the probit model. The file emm_moments.m defines EMM moment conditions, where the DGP and score generator can be passed as arguments. Thus, it is a general purpose moment condition for EMM estimation. This file is interesting enough to warrant some discussion. A listing appears in Listing 19.1. Line 3 defines the DGP, and the arguments needed to evaluate it are defined in line 4. The score generator is defined in line 5, and its arguments are defined in line 6. The QML estimate of the parameter of the score generator is read in line 7. Note in line 10 how the random draws needed to simulate data are passed with the data, and are thus fixed during estimation, to avoid ”chattering”. The simulated data is generated in line 16, and the derivative of the score generator using the simulated data is calculated in line 18. In line 20 we average the scores of the score generator, which are the moment conditions that the function returns.

1

function scores = emm_moments(theta, data, momentargs)

2

k = momentargs{1};

3

dgp = momentargs{2}; # the data generating process (DGP)

4

dgpargs = momentargs{3}; # its arguments (cell array)

5

sg = momentargs{4}; # the score generator (SG)

6

sgargs = momentargs{5}; # SG arguments (cell array)

7

phi = momentargs{6}; # QML estimate of SG parameter

8

y = data(:,1);

9

x = data(:,2:k+1);

10

rand_draws = data(:,k+2:columns(data)); # passed with data to ensure fixed across iterations

19.5. EXAMPLES 11

n = rows(y);

12

scores = zeros(n,rows(phi)); # container for moment contributions

13

reps = columns(rand_draws); # how many simulations?

14

for i = 1:reps

423

15

e = rand_draws(:,i);

16

y = feval(dgp, theta, x, e, dgpargs); # simulated data

17

sgdata = [y x]; # simulated data for SG

18

scores = scores + numgradient(sg, {phi, sgdata, sgargs}); # gradient of SG

19

endfor

20

scores = scores / reps; # average over number of simulations

21

endfunction

L ISTING 19.1 The file emm_example.m performs EMM estimation of the probit model, using a logit model as the score generator. The results we obtain are Score generator results: ===================================================== BFGSMIN final results Used analytic gradient -----------------------------------------------------STRONG CONVERGENCE Function conv 1 Param conv 1 Gradient conv 1 -----------------------------------------------------Objective function value 0.281571 Stepsize 0.0279 15 iterations -----------------------------------------------------param

gradient

change

19.5. EXAMPLES

1.8979 1.6648 1.9125 1.8875 1.7433

0.0000 -0.0000 -0.0000 -0.0000 -0.0000

424

0.0000 0.0000 0.0000 0.0000 0.0000

====================================================== Model results: ****************************************************** EMM example GMM Estimation Results BFGS convergence: Normal convergence Objective function value: 0.000000 Observations: 1000 Exactly identified, no spec. test estimate st. err t-stat p-value p1 1.069 0.022 47.618 0.000 p2 0.935 0.022 42.240 0.000 p3 1.085 0.022 49.630 0.000 p4 1.080 0.022 49.047 0.000 p5 0.978 0.023 41.643 0.000 ****************************************************** It might be interesting to compare the standard errors with those obtained from ML estimation, to check efficiency of the EMM estimator. One could even do a Monte Carlo study.

19.5. EXAMPLES

425

Exercises (1) Do SML estimation of the probit model. (2) Do a little Monte Carlo study to compare ML, SML and EMM estimation of the probit model. Investigate how the number of simulations affect the two simulation-based estimators.

CHAPTER 20

Parallel programming for econometrics The following borrows heavily from Creel (2005). Parallel computing can offer an important reduction in the time to complete computations. This is well-known, but it bears emphasis since it is the main reason that parallel computing may be attractive to users. To illustrate, the Intel Pentium IV (Willamette) processor, running at 1.5GHz, was introduced in November of 2000. The Pentium IV (Northwood-HT) processor, running at 3.06GHz, was introduced in November of 2002. An approximate doubling of the performance of a commodity CPU took place in two years. Extrapolating this admittedly rough snapshot of the evolution of the performance of commodity processors, one would need to wait more than 6.6 years and then purchase a new computer to obtain a 10-fold improvement in computational performance. The examples in this chapter show that a 10-fold improvement in performance can be achieved immediately, using distributed parallel computing on available computers. Recent (this is written in 2005) developments that may make parallel computing attractive to a broader spectrum of researchers who do computations. The first is the fact that setting up a cluster of computers for distributed parallel computing is not difficult. If you are using the ParallelKnoppix bootable CD that accompanies these notes, you could create a cluster in less that 10 minutes, given a few easily satisfied conditions. See the Tutorial, on the Desktop if you’re running the bootable CD, or available at the preceeding link. The second development is the existence of extensions to some of

426

20.1. EXAMPLE PROBLEMS

427

the high-level matrix programming (HLMP) languages1 that allow the incorporation of parallelism into programs written in these languages. Following are examples of parallel implementations of several mainstream problems in econometrics. A focus of the examples is on the possibility of hiding parallelization from end users of programs. If programs that run in parallel have an interface that is nearly identical to the interface of equivalent serial versions, end users will find it easy to take advantage of parallel computing’s performance. We continue to use Octave, taking advantage of the MPI Toolbox (MPITB) for Octave, by by Fernández Baldomero et al. (2004). There are also parallel packages for Ox, R, and Python which may be of interest to econometricians, but as of this writing, the following examples are the most accessible introduction to parallel programming for econometricians. 20.1. Example problems This section introduces example problems from econometrics, and shows how they can be parallelized in a natural way. 20.1.1. Monte Carlo. A Monte Carlo study involves repeating a random experiment many times under identical conditions. Several authors have noted that Monte Carlo studies are obvious candidates for parallelization (Doornik et al. 2002; Bruche, 2003) since blocks of replications can be done independently on different computers. To illustrate the parallelization of a Monte Carlo study, we use same trace test example as do Doornik, et. al. (2002). tracetest.m is a function that calculates the trace test statistic for the lack of cointegration of integrated time series. This function is illustrative of the format that we adopt for Monte Carlo simulation of a function: it receives a single argument of cell type, and it returns a row vector that holds the results of 1By ”high-level matrix programming language” I mean languages such as MATLAB (TM the Math-

works, Inc.), Ox (TM OxMetrics Technologies, Ltd.), and GNU Octave (www.octave.org), for example.


428

one random simulation. The single argument in this case is a cell array that holds the length of the series in its first position, and the number of series in the second position. It generates a random result though a process that is internal to the function, and it reports some output in a row vector (in this case the result is a scalar). mc_example1.m is an Octave script that executes a Monte Carlo study of the trace test by repeatedly evaluating the tracetest.m function. The main thing to notice about this script is that lines 7 and 10 call the function montecarlo.m. When called with 3 arguments, as in line 7, montecarlo.m executes serially on the computer it is called from. In line 10, there is a fourth argument. When called with four arguments, the last argument is the number of slave hosts to use. We see that running the Monte Carlo study on one or more processors is transparent to the user - he or she must only indicate the number of slave computers to be used. 20.1.2. ML. For a sample {(yt , xt )}n of n observations of a set of dependent and

explanatory variables, the maximum likelihood estimator of the parameter θ can be defined as θˆ = arg max sn (θ) where sn (θ) =

1 n ∑ ln f (yt |xt , θ) n t=1

Here, yt may be a vector of random variables, and the model may be dynamic since xt may contain lags of yt . As Swann (2002) points out, this can be broken into sums over blocks of observations, for example two blocks: ( ! n1 1 sn (θ) = ∑ ln f (yt |xt , θ) + n t=1

n

∑

t=n1 +1

ln f (yt |xt , θ)

!)

Analogously, we can define up to n blocks. Again following Swann, parallelization can be done by calculating each block on separate computers.


429

mle_example1.m is an Octave script that calculates the maximum likelihood estimator of the parameter vector of a model that assumes that the dependent variable is distributed as a Poisson random variable, conditional on some explanatory variables. In lines 1-3 the data is read, the name of the density function is provided in the variable model, and the initial value of the parameter vector is set. In line 5, the function mle_estimate performs ordinary serial calculation of the ML estimator, while in line 7 the same function is called with 6 arguments. The fourth and fifth arguments are empty placeholders where options to mle_estimate may be set, while the sixth argument is the number of slave computers to use for parallel execution, 1 in this case. A person who runs the program sees no parallel programming code - the parallelization is transparent to the end user, beyond having to select the number of slave computers. When executed, this script prints out the estimates theta_s and theta_p, which are identical. It is worth noting that a different likelihood function may be used by making the model variable point to a different function. The likelihood function itself is an ordinary Octave function that is not parallelized. The mle_estimate function is a generic function that can call any likelihood function that has the appropriate input/output syntax for evaluation either serially or in parallel. Users need only learn how to write the likelihood function using the Octave language.

20.1.3. GMM. For a sample as above, the GMM estimator of the parameter θ can be defined as θˆ ≡ arg min sn (θ) Θ

where sn (θ) = mn (θ)0Wn mn (θ)


and mn (θ) =

430

1 n ∑ mt (yt |xt , θ) n t=1

Since mn (θ) is an average, it can obviously be computed blockwise, using for example 2 blocks: (20.1.1)

1 mn (θ) = n

(

n1

∑ mt (yt |xt , θ)

t=1

!

n

+

∑

t=n1 +1

mt (yt |xt , θ)

!)

Likewise, we may define up to n blocks, each of which could potentially be computed on a different machine. gmm_example1.m is a script that illustrates how GMM estimation may be done serially or in parallel. When this is run, theta_s and theta_p are identical up to the tolerance for convergence of the minimization routine. The point to notice here is that an end user can perform the estimation in parallel in virtually the same way as it is done serially. Again, gmm_estimate, used in lines 8 and 10, is a generic function that will estimate any model specified by the moments variable - a different model can be estimated by changing the value of the moments variable. The function that moments points to is an ordinary Octave function that uses no parallel programming, so users can write their models using the simple and intuitive HLMP syntax of Octave. Whether estimation is done in parallel or serially depends only the seventh argument to gmm_estimate - when it is missing or zero, estimation is by default done serially with one processor. When it is positive, it specifies the number of slave nodes to use.


431

20.1.4. Kernel regression. The Nadaraya-Watson kernel regression estimator of a function g(x) at a point x is g(x) ˆ =

n yt K [(x − xt ) /γn ] ∑t=1 n ∑t=1 K [(x − xt ) /γn ] n

≡

∑ wt yy

t=1

We see that the weight depends upon every data point in the sample. To calculate the fit at every point in a sample of size n, on the order of n2 k calculations must be done, where k is the dimension of the vector of explanatory variables, x. Racine (2002) demonstrates that MPI parallelization can be used to speed up calculation of the kernel regression estimator by calculating the fits for portions of the sample on different computers. We follow this implementation here. kernel_example1.m is a script for serial and parallel kernel regression. Serial execution is obtained by setting the number of slaves equal to zero, in line 15. In line 17, a single slave is specified, so execution is in parallel on the master and slave nodes. The example programs show that parallelization may be mostly hidden from end users. Users can benefit from parallelization without having to write or understand parallel code. The speedups one can obtain are highly dependent upon the specific problem at hand, as well as the size of the cluster, the efficiency of the network, etc. Some examples of speedups are presented in Creel (2005). Figure 20.1.1 reproduces speedups for some econometric problems on a cluster of 12 desktop computers. The speedup for k nodes is the time to finish the problem on a single node divided by the time to finish the problem on k nodes. Note that you can get 10X speedups, as claimed in the introduction. It’s pretty obvious that much greater speedups could be obtained using a larger cluster, for the ”embarrassingly parallel” problems.


432

F IGURE 20.1.1. Speedups from parallelization 11 MONTECARLO BOOTSTRAP MLE GMM KERNEL

10 9 8 7 6 5 4 3 2 1

2

4

6

8 nodes

10

12

Bibliography [1] Bruche, M. (2003) A note on embarassingly parallel computation using OpenMosix and Ox, working paper, Financial Markets Group, London School of Economics. [2] Creel, M. (2005) User-friendly parallel computations with econometric examples, Computational Economics, V. 26, pp. 107-128. [3] Doornik, J.A., D.F. Hendry and N. Shephard (2002) Computationally-intensive econometrics using a distributed matrix-programming language, Philosophical Transactions of the Royal Society of London, Series A, 360, 1245-1266. [4] Fernández Baldomero, J. (2004) LAM/MPI parallel computing under GNU Octave, atc.ugr.es/javier-bin/mpitb. [5] Racine, Jeff (2002) Parallel distributed kernel estimation, Computational Statistics & Data Analysis, 40, 293-302. [6] Swann, C.A. (2002) Maximum likelihood estimation using parallel computing: an introduction to MPI, Computational Economics, 19, 145-178.

433

CHAPTER 21

Final project: econometric estimation of a RBC model THIS IS NOT FINISHED - IGNORE IT FOR NOW In this last chapter we’ll go through a worked example that combines a number of the topics we’ve seen. We’ll do simulated method of moments estimation of a real business cycle model, similar to what Valderrama (2002) does. 21.1. Data We’ll develop a model for private consumption and real gross private investment. The data are obtained from the US Bureau of Economic Analysis (BEA) National Income and Product Accounts (NIPA), Table 11.1.5, Lines 2 and 6 (you can download quarterly data from 1947-I to the present). The data we use are in the file rbc_data.m. This data is real (constant dollars). The program plots.m will make a few plots, including Figures 21.1.1 though 21.1.3. First looking at the plot for levels, we can see that real consumption and investment are clearly nonstationary (surprise, surprise). There appears to be somewhat of a structural change in the mid-1970’s. F IGURE 21.1.1. Consumption and Investment, Levels

Include/RBC/levels.eps not found!

434

21.1. DATA

435

F IGURE 21.1.2. Consumption and Investment, Growth Rates

Include/RBC/growth.eps not found!

F IGURE 21.1.3. Consumption and Investment, Bandpass Filtered

Include/RBC/filtered.eps not found!

Looking at growth rates, the series for consumption has an extended period of high growth in the 1970’s, becoming more moderate in the 90’s. The volatility of growth of consumption has declined somewhat, over time. Looking at investment, there are some notable periods of high volatility in the mid-1970’s and early 1980’s, for example. Since 1990 or so, volatility seems to have declined. Economic models for growth often imply that there is no long term growth (!) - the data that the models generate is stationary and ergodic. Or, the data that the models generate needs to be passed through the inverse of a filter. We’ll follow this, and generate stationary business cycle data by applying the bandpass filter of Christiano and Fitzgerald (1999). The filtered data is in Figure 21.1.3. We’ll try to specify an economic model that can generate similar data. To get data that look like the levels for consumption and investment, we’d need to apply the inverse of the bandpass filter.

21.2. AN RBC MODEL

436

21.2. An RBC Model Consider a very simple stochastic growth model (the same used by Maliar and Maliar (2003), with minor notational difference): ∞ E0 ∑∞ βt U (ct ) max{ct ,kt }t=0 t=0

ct + kt

=

α (1 − δ) kt−1 + φt kt−1

log φt

=

ρ log φt−1 + εt

εt

∼

IIN(0, σ2ε )

Assume that the utility function is 1−γ

U (ct ) =

ct

−1 1−γ

• β is the discount rate

• δ is the depreciation rate of capital

• α is the elasticity of output with respect to capital

• φ is a technology shock that is positive. φt is observed in period t.

• γ is the coefficient of relative risk aversion. When γ = 1, the utility function is logarithmic. • gross investment, it , is the change in the capital stock: it = kt − (1 − δ) kt−1 • we assume that the initial condition (k0 , θ0 ) is given. We would like to estimate the parameters θ = β, γ, δ, α, ρ, σ2ε

0

using the data that

we have on consumption and investment. This problem is very similar to the GMM estimation of the portfolio model discussed in Sections 15.11 and 15.12. Once can

21.3. A REDUCED FORM MODEL

437

derive the Euler condition in the same way we did there, and use it to define a GMM estimator. That approach was not very successful, recall. Now we’ll try to use some more informative moment conditions to see if we get better results. 21.3. A reduced form model Macroeconomic time series data are often modeled using vector autoregressions. A vector autogression is just the vector version of an autoregressive model. Let yt be a G-vector of jointly dependent variables. A VAR(p) model is yt = c + A1 yt−1 + A2 yt−2 + ... + A p yt−p + vt where c is a G-vector of parameters, and A j , j=1,2,...,p, are G × G matrices of parame-

ters. Let vt = Rt ηt , where ηt ∼ IIN(0, I2), and Rt is upper triangular. So V (vt |yt−1 , ...yt−p ) = 0

Rt Rt . You can think of a VAR model as the reduced form of a dynamic linear simultaneous equations model where all of the variables are treated as endogenous. Clearly, if all of the variables are endogenous, one would need some form of additional information to identify a structural model. But we already have a structural model, and we’re only going to use the VAR to help us estimate the parameters. A well-fitting reduced form model will be adequate for the purpose. We’re seen that our data seems to have episodes where the variance of growth rates and filtered data is non-constant. This brings us to the general area of stochastic volatility. Without going into details, we’ll just consider the exponential GARCH model of Nelson (1991) as presented in Hamilton (1994, pg. 668-669). Define ht = vec∗ (Rt ), the vector of elements in the upper triangle of Rt (in our case this is a 3 × 1 vector). We assume that the elements follow o n p log h jt = κ j + P( j,.) |vt−1 | − 2/π + ℵ( j,.) vt−1 + G( j,.) log ht−1

21.3. A REDUCED FORM MODEL

438

The variance of the VAR error depends upon its own past, as well as upon the past realizations of the shocks.

• This is an EGARCH(1,1) specification. The obvious generalization is the EGARCH(r, m) specification, with longer lags (r for lags of v, m for lags of h). • The advantage of the EGARCH formulation is that the variance is assuredly positive without parameter restrictions • The matrix P has dimension 3 × 2. • The matrix G has dimension 3 × 3.

• The matrix ℵ (reminder to self: this is an ”aleph”) has dimension 2 × 2.

• The parameter matrix ℵ allows for leverage, so that positive and negative shocks can have asymmetric effects upon volatility. • We will probably want to restrict these parameter matrices in some way. For instance, G could plausibly be diagonal.

With the above specification, we have ηt ∼ IIN (0, I2 ) ηt = Rt−1 vt and we know how to calculate Rt and vt , given the data and the parameters. Thus, it is straighforward to do estimation by maximum likelihood. This will be the score generator.

21.5. SOLVING THE STRUCTURAL MODEL

439

21.4. Results (I): The score generator 21.5. Solving the structural model The first order condition for the structural model is −γ ct

or ct =

= βEt

n

−γ ct+1

h

−γ βEt ct+1

1 − δ + αφt+1 ktα−1

1 − δ + αφt+1 ktα−1

io −1γ

The problem is that we cannot solve for ct since we do not know the solution for the expectation in the previous equation. The parameterized expectations algorithm (PEA: den Haan and Marcet, 1990), is a means of solving the problem. The expectations term is replaced by a parametric function. As long as the parametric function is a flexible enough function of variables that have been realized in period t, there exist parameter values that make the approximation as close to the true expectation as is desired. We will write the approximation h

−γ Et ct+1

1 − δ + αφt+1 ktα−1

i

' exp (ρ0 + ρ1 log φt + ρ2 log kt−1 )

For given values of the parameters of this approximating function, we can solve for ct , and then for kt using the restriction that α ct + kt = (1 − δ) kt−1 + φt kt−1

This allows us to generate a series {(ct , kt )}. Then the expectations approximation is updated by fitting −γ ct+1 1 − δ + αφt+1 ktα−1 = exp (ρ0 + ρ1 log φt + ρ2 log kt−1 ) + ηt

21.5. SOLVING THE STRUCTURAL MODEL

440

by nonlinear least squares. The 2 step procedure of generating data and updating the parameters of the approximation to expectations is iterated until the parameters no longer change. When this is the case, the expectations function is the best fit to the generated data. As long it is a rich enough parametric model to encompass the true expectations function, it can be made to be equal to the true expectations function by using a long enough simulation. 0 Thus, given the parameters of the structural model, θ = β, γ, δ, α, ρ, σ2ε , we can

generate data {(ct , kt )} using the PEA. From this we can get the series {(ct , it )} using it = kt − (1 − δ) kt−1 . This can be used to do EMM estimation using the scores of the

reduced form model to define moments, using the simulated data from the structural model.

Bibliography [1] Creel. M (2005) A Note on Parallelizing the Parameterized Expectations Algorithm. [2] den Haan, W. and Marcet, A. (1990) Solving the stochastic growth model by parameterized expectations, Journal of Business and Economics Statistics, 8, 31-34. [3] Hamilton, J. (1994) Time Series Analysis, Princeton Univ. Press

[4] Maliar, L. and Maliar, S. (2003) Matlab code for Solving a Neoclassical Growh Model with a Parametrized Expectations A [5] Nelson, D. (1991) Conditional heteroscedasticity is asset returns: a new approach, Econometrica, 59, 347-70. [6] Valderrama, D. (2002) Statistical nonlinearities in the business cycle:

a challenge for

the canonical RBC model, Economic Research, Federal Reserve Bank of San Francisco. http://ideas.repec.org/p/fip/fedfap/2002-13.html

441

CHAPTER 22

Introduction to Octave Why is Octave being used here, since it’s not that well-known by econometricians? Well, because it is a high quality environment that is easily extensible, uses well-tested and high performance numerical libraries, it is licensed under the GNU GPL, so you can get it for free and modify it if you like, and it runs on both GNU/Linux, Mac OSX and Windows systems. It’s also quite easy to learn. 22.1. Getting started Get the bootable CD, as was described in Section 1.3. Then burn the image, and boot your computer with it. This will give you this same PDF file, but with all of the example programs ready to run. The editor is configure with a macro to execute the programs using Octave, which is of course installed. From this point, I assume you are running the CD (or sitting in the computer room across the hall from my office), or that you have configured your computer to be able to run the *.m files mentioned below. 22.2. A short introduction The objective of this introduction is to learn just the basics of Octave. There are other ways to use Octave, which I encourage you to explore. These are just some rudiments. After this, you can look at the example programs scattered throughout the document (and edit them, and run them) to learn more about how Octave can be used to do econometrics. Students of mine: your problem sets will include exercises that 442

22.2. A SHORT INTRODUCTION

443

F IGURE 22.2.1. Running an Octave program

can be done by modifying the example programs in relatively minor ways. So study the examples! Octave can be used interactively, or it can be used to run programs that are written using a text editor. We’ll use this second method, preparing programs with NEdit, and calling Octave from within the editor. The program first.m gets us started. To run this, open it up with NEdit (by finding the correct file inside the /home/knoppix/Desktop/Econometrics folder and clicking on the icon) and then type CTRL-ALT-o, or use the Octave item in the Shell menu (see Figure 22.2.1).

22.3. IF YOU’RE RUNNING A LINUX INSTALLATION...

444

Note that the output is not formatted in a pleasing way. That’s because printf() doesn’t automatically start a new line. Edit first.m so that the 8th line reads ”printf(”hello world\n”);” and re-run the program. We need to know how to load and save data. The program second.m shows how. Once you have run this, you will find the file ”x” in the directory Econometrics/Include/OctaveIntro/ You might have a look at it with NEdit to see Octave’s default format for saving data. Basically, if you have data in an ASCII text file, named for example ”myfile.data”, formed of numbers separated by spaces, just use the command ”load myfile.data”. After having done so, the matrix ”myfile” (without extension) will contain the data. Please have a look at CommonOperations.m for examples of how to do some basic things in Octave. Now that we’re done with the basics, have a look at the Octave programs that are included as examples. If you are looking at the browsable PDF version of this document, then you should be able to click on links to open them. If not, the example programs are available here and the support files needed to run these are available here. Those pages will allow you to examine individual files, out of context. To actually use these files (edit and run them), you should go to the home page of this document, since you will probably want to download the pdf version together with all the support files and examples. Or get the bootable CD. There are some other resources for doing econometrics with Octave. You might like to check the article Econometrics with Octave and the Econometrics Toolbox , which is for Matlab, but much of which could be easily used with Octave. 22.3. If you’re running a Linux installation... Then to get the same behavior as found on the CD, you need to: • Get the collection of support programs and the examples, from the document home page.

22.3. IF YOU’RE RUNNING A LINUX INSTALLATION...

445

• Put them somewhere, and tell Octave how to find them, e.g., by putting a link to the MyOctaveFiles directory in /usr/local/share/octave/site-m • Make sure nedit is installed and configured to run Octave and use syntax highlighting. Copy the file /home/econometrics/.nedit from the CD to do this. Or, get the file NeditConfiguration and save it in your $HOME directory with the name ”.nedit”. Not to put too fine a point on it, please note that there is a period in that name. • Associate *.m files with NEdit so that they open up in the editor when you click on them. That should do it.

CHAPTER 23

Notation and Review • All vectors will be column vectors, unless they have a transpose symbol (or I forget to apply this rule - your help catching typos and er0rors is much appreciated). For example, if xt is a p × 1 vector, xt0 is a 1 × p vector. When I refer to a p-vector, I mean a column vector.

23.1. Notation for differentiation of vectors and matrices [3, Chapter 1] Let s(·) : ℜ p → ℜ be a real valued function of the p-vector θ. Then nized as a p-vector,



  ∂s(θ)   =  ∂θ  

∂s(θ) ∂θ1 ∂s(θ) ∂θ2

.. .

∂s(θ) ∂θ p

∂s(θ) ∂θ0

is orga-

        

Following this convention, ∂s(θ) ∂θ0 is a 1 × p vector, and ∂2 s(θ) ∂ = 0 ∂θ∂θ ∂θ

∂s(θ) ∂θ

∂ = 0 ∂θ

∂2 s(θ) ∂θ∂θ0

is a p × p matrix. Also,

∂s(θ) . ∂θ

E XERCISE 33. For a and x both p-vectors, show that

∂a0 x ∂x

= a.

Let f (θ):ℜ p → ℜn be a n-vector valued function of the p-vector θ. Let f (θ)0 be 0 ∂ f (θ)0 = ∂θ∂ 0 f (θ). the 1 × n valued transpose of f . Then ∂θ 446

23.2. CONVERGENGE MODES

447

• Product rule: Let f (θ):ℜ p → ℜn and h(θ):ℜ p → ℜn be n-vector valued functions of the p-vector θ. Then

∂ ∂ ∂ 0 0 0 h(θ) f (θ) = h f +f h ∂θ0 ∂θ0 ∂θ0 has dimension 1 × p. Applying the transposition rule we get ∂ 0 ∂ ∂ 0 0 h(θ) f (θ) = f h+ h f ∂θ ∂θ ∂θ which has dimension p × 1. E XERCISE 34. For A a p × p matrix and x a p × 1 vector, show that

∂x0 Ax ∂x

= A + A0 .

• Chain rule: Let f (·):ℜ p → ℜn a n-vector valued function of a p-vector ar-

gument, and let g():ℜr → ℜ p be a p-vector valued function of an r-vector

valued argument ρ. Then

has dimension n × r.

∂ ∂ ∂ f [g (ρ)] = f (θ) g(ρ) 0 0 0 ∂ρ ∂θ θ=g(ρ) ∂ρ

E XERCISE 35. For x and β both p × 1 vectors, show that

∂ exp(x0 β) ∂β

= exp(x0 β)x.

23.2. Convergenge modes Readings: [1, Chapter 4];[4, Chapter 4]. We will consider several modes of convergence. The first three modes discussed are simply for background. The stochastic modes are those which will be used later in the course. D EFINITION 36. A sequence is a mapping from the natural numbers {1, 2, ...} =

{n}∞ n=1 = {n} to some other set, so that the set is ordered according to the natural numbers associated with its elements.


448

Real-valued sequences: D EFINITION 37. [Convergence] A real-valued sequence of vectors {a n } converges

to the vector a if for any ε > 0 there exists an integer Nε such that for all n > Nε , k an − a k< ε . a is the limit of an , written an → a.

Deterministic real-valued functions. Consider a sequence of functions { f n (ω)} where fn : Ω → T ⊆ ℜ. Ω may be an arbitrary set. D EFINITION 38. [Pointwise convergence] A sequence of functions { f n (ω)} con-

verges pointwise on Ω to the function f (ω) if for all ε > 0 and ω ∈ Ω there exists an integer Nεω such that | fn (ω) − f (ω)| < ε, ∀n > Nεω .

It’s important to note that Nεω depends upon ω, so that converge may be much more rapid for certain ω than for others. Uniform convergence requires a similar rate of convergence throughout Ω. D EFINITION 39. [Uniform convergence] A sequence of functions { f n (ω)} con-

verges uniformly on Ω to the function f (ω) if for any ε > 0 there exists an integer N such that sup | fn (ω) − f (ω)| < ε, ∀n > N.

ω∈Ω

(insert a diagram here showing the envelope around f (ω) in which f n (ω) must lie) Stochastic sequences. In econometrics, we typically deal with stochastic sequences. Given a probability space (Ω, F , P) , recall that a random variable maps the sample space to the real line, i.e., X (ω) : Ω → ℜ. A sequence of random variables {Xn (ω)} is


449

a collection of such mappings, i.e., each Xn (ω) is a random variable with respect to the probability space (Ω, F , P) . For example, given the model Y = X β0 + ε, the OLS estimator βˆ n = (X 0 X )−1 X 0Y, where n is the sample size, can be used to form a sequence of random vectors {βˆ n }. A number of modes of convergence are in use when dealing with sequences of random variables. Several such modes of convergence should already be familiar: D EFINITION 40. [Convergence in probability] Let Xn (ω) be a sequence of random variables, and let X (ω) be a random variable. Let An = {ω : |Xn (ω)−X (ω)| > ε}. Then {Xn (ω)} converges in probability to X (ω) if lim P (An ) = 0, ∀ε > 0.

n→∞

p

Convergence in probability is written as Xn → X , or plim Xn = X . D EFINITION 41. [Almost sure convergence] Let Xn (ω) be a sequence of random variables, and let X (ω) be a random variable. Let A = {ω : limn→∞ Xn (ω) = X (ω)}. Then {Xn (ω)} converges almost surely to X (ω) if P (A ) = 1. In other words, Xn (ω) → X (ω) (ordinary convergence of the two functions) except on a.s.

a set C = Ω − A such that P(C) = 0. Almost sure convergence is written as Xn → X , or Xn → X , a.s. One can show that a.s.

p

Xn → X ⇒ Xn → X . D EFINITION 42. [Convergence in distribution] Let the r.v. Xn have distribution function Fn and the r.v. Xn have distribution function F. If Fn → F at every continuity point of F, then Xn converges in distribution to X .


450

d

Convergence in distribution is written as Xn → X . It can be shown that convergence in probability implies convergence in distribution. Stochastic functions. Simple laws of large numbers (LLN’s) allow us to directly a.s. conclude that βˆ n → β0 in the OLS example, since

0 −1 0 Xε ˆβn = β0 + X X , n n

and

X 0ε n

a.s.

→ 0 by a SLLN. Note that this term is not a function of the parameter β.

This easy proof is a result of the linearity of the model, which allows us to express the estimator in a way that separates parameters from random functions. In general, this is not possible. We often deal with the more complicated situation where the stochastic sequence depends on parameters in a manner that is not reducible to a simple sequence of random variables. In this case, we have a sequence of random functions that depend on θ: {Xn (ω, θ)}, where each Xn (ω, θ) is a random variable with respect to a probability space (Ω, F , P) and the parameter θ belongs to a parameter space θ ∈ Θ.

D EFINITION 43. [Uniform almost sure convergence] {Xn (ω, θ)} converges uni-

formly almost surely in Θ to X (ω, θ) if

lim sup |Xn (ω, θ) − X (ω, θ)| = 0, (a.s.)

n→∞ θ∈Θ

Implicit is the assumption that all Xn (ω, θ) and X (ω, θ) are random variables w.r.t. u.a.s.

(Ω, F , P) for all θ ∈ Θ. We’ll indicate uniform almost sure convergence by → and u.p.

uniform convergence in probability by → . • An equivalent definition, based on the fact that “almost sure” means “with probability one” is Pr

lim sup |Xn (ω, θ) − X (ω, θ)| = 0 = 1

n→∞ θ∈Θ

23.3. RATES OF CONVERGENCE AND ASYMPTOTIC EQUALITY

451

This has a form similar to that of the definition of a.s. convergence - the essential difference is the addition of the sup. 23.3. Rates of convergence and asymptotic equality It’s often useful to have notation for the relative magnitudes of quantities. Quantities that are small relative to others can often be ignored, which simplifies analysis. D EFINITION 44. [Little-o] Let f (n) and g(n) be two real-valued functions. The f (n)

notation f (n) = o(g(n)) means limn→∞ g(n) = 0. D EFINITION 45. [Big-O] Let f (n) and g(n) be two real-valued functions. The f (n) notation f (n) = O(g(n)) means there exists some N such that for n > N, g(n) < K,

where K is a finite constant.

This definition doesn’t require that

f (n) g(n)

have a limit (it may fluctuate boundedly).

If { fn } and {gn } are sequences of random variables analogous definitions are D EFINITION 46. The notation f (n) = o p (g(n)) means

f (n) p g(n) → 0.

E XAMPLE 47. The least squares estimator θˆ = (X 0 X )−1 X 0Y = (X 0 X )−1 X 0 X θ0 + ε = 0

−1 X 0 ε

θ0 + (X 0 X )−1 X 0 ε. Since plim (X X)1

= 0, we can write (X 0X )−1 X 0 ε = o p (1) and

θˆ = θ0 + o p (1). Asymptotically, the term o p (1) is negligible. This is just a way of indicating that the LS estimator is consistent. D EFINITION 48. The notation f (n) = O p (g(n)) means there exists some Nε such that for ε > 0 and all n > Nε ,

where Kε is a finite constant.

f (n) < Kε > 1 − ε, P g(n)

23.3. RATES OF CONVERGENCE AND ASYMPTOTIC EQUALITY

452

E XAMPLE 49. If Xn ∼ N(0, 1) then Xn = O p (1), since, given ε, there is always

some Kε such that P (|Xn | < Kε ) > 1 − ε. Useful rules: • O p (n p )O p (nq ) = O p (n p+q ) • o p (n p )o p (nq ) = o p (n p+q )

E XAMPLE 50. Consider a random sample of iid r.v.’s with mean 0 and variance σ2 . The estimator of the mean θˆ = 1/n ∑ni=1 xi is asymptotically normally distributed, A e.g., n1/2 θˆ ∼ N(0, σ2 ). So n1/2 θˆ = O p (1), so θˆ = O p (n−1/2 ). Before we had θˆ = o p (1),

now we have have the stronger result that relates the rate of convergence to the sample size. E XAMPLE 51. Now consider a random sample of iid r.v.’s with mean µ and variance σ2 . The estimator of the mean θˆ = 1/n ∑ni=1 xi is asymptotically normally dis A tributed, e.g., n1/2 θˆ − µ ∼ N(0, σ2 ). So n1/2 θˆ − µ = O p (1), so θˆ − µ = O p (n−1/2 ), so θˆ = O p (1).

These two examples show that averages of centered (mean zero) quantities typically have plim 0, while averages of uncentered quantities have finite nonzero plims. Note that the definition of O p does not mean that f (n) and g(n) are of the same order. Asymptotic equality ensures that this is the case. D EFINITION 52. Two sequences of random variables { f n } and {gn } are asymptota

ically equal (written fn = gn ) if

f (n) plim g(n)

=1

Finally, analogous almost sure versions of o p and O p are defined in the obvious way.

EXERCISES

453

Exercises (1) For a and x both p × 1 vectors, show that

∂a0 x ∂x

= a.

(2) For A a p × p matrix and x a p × 1 vector, show that

∂x0 Ax ∂x

= A + A0 .

(3) For x and β both p × 1 vectors, show that Dβ exp x0 β = exp(x0 β)x.

(4) For x and β both p × 1 vectors, find the analytic expression for D 2β exp x0 β. (5) Write an Octave program that verifies each of the previous results by taking numeric derivatives. For a hint, type help numgradient and help numhessian inside octave.

CHAPTER 24

The GPL This document and the associated examples and materials are copyright Michael Creel, under the terms of the GNU General Public License. This license follows: GNU GENERAL PUBLIC LICENSE Version 2, June 1991 Copyright (C) 1989, 1991 Free Software Foundation, Inc. 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. Preamble The licenses for most software are designed to take away your freedom to share and change it. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software–to make sure the software is free for all its users. This General Public License applies to most of the Free Software Foundation’s software and to any other program whose authors commit to using it. (Some other Free Software Foundation software is covered by the GNU Library General Public License instead.) You can apply it to your programs, too. When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things.

454

24. THE GPL

455

To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it. For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights. We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which gives you legal permission to copy, distribute and/or modify the software. Also, for each author’s protection and ours, we want to make certain that everyone understands that there is no warranty for this free software. If the software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors’ reputations. Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone’s free use or not licensed at all. The precise terms and conditions for copying, distribution and modification follow. GNU GENERAL PUBLIC LICENSE TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION 0. This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work

24. THE GPL

456

under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, translation is included without limitation in the term "modification".) Each licensee is addressed as "you". Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does. 1. You may copy and distribute verbatim copies of the Program’s source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program. You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. 2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: a) You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change. b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.

24. THE GPL

457

c) If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is interactive but does not normally print such an announcement, your work based on the Program is not required to print an announcement.) These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it. Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program. In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License. 3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:

24. THE GPL

458

a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.) The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code.

24. THE GPL

459

4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. 5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it. 6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients’ exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License. 7. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then

24. THE GPL

460

the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program. If any portion of this section is held invalid or unenforceable under any particular circumstance, the balance of the section is intended to apply and the section as a whole is intended to apply in other circumstances. It is not the purpose of this section to induce you to infringe any patents or other property right claims or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of the free software distribution system, which is implemented by public license practices. Many people have made generous contributions to the wide range of software distributed through that system in reliance on consistent application of that system; it is up to the author/donor to decide if he or she is willing to distribute software through any other system and a licensee cannot impose that choice. This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License. 8. If the distribution and/or use of the Program is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License. 9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later

24. THE GPL

461

version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation. 10. If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally. NO WARRANTY 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT

24. THE GPL

462

NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. END OF TERMS AND CONDITIONS How to Apply These Terms to Your New Programs If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms. To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the "copyright" line and a pointer to where the full notice is found. Copyright (C) 19yy This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA

24. THE GPL

463

Also add information on how to contact you by electronic and paper mail. If the program is interactive, make it output a short notice like this when it starts in an interactive mode: Gnomovision version 69, Copyright (C) 19yy name of author Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type ‘show w’. This is free software, and you are welcome to redistribute it under certain conditions; type ‘show c’ for details. The hypothetical commands ‘show w’ and ‘show c’ should show the appropriate parts of the General Public License. Of course, the commands you use may be called something other than ‘show w’ and ‘show c’; they could even be mouse-clicks or menu items–whatever suits your program. You should also get your employer (if you work as a programmer) or your school, if any, to sign a "copyright disclaimer" for the program, if necessary. Here is a sample; alter the names: Yoyodyne, Inc., hereby disclaims all copyright interest in the program ‘Gnomovision’ (which makes passes at compilers) written by James Hacker. <signature of Ty Coon>, 1 April 1989 Ty Coon, President of Vice This General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Library General Public License instead of this License.

CHAPTER 25

The attic This holds material that is not really ready to be incorporated into the main body, but that I don’t want to lose. Basically, ignore it, unless you’d like to help get it ready for inclusion. 25.1. Hurdle models Returning to the Poisson model, lets look at actual and fitted count probabilities. Actual relative frequencies are f (y = j) = ∑i 1(yi = j)/n and fitted frequencies are ˆ fˆ(y = j) = ∑ni=1 fY ( j|xi , θ)/n We see that for the OBDV measure, there are many TABLE 1. Actual and Poisson fitted frequencies Count OBDV ERV Count Actual Fitted Actual Fitted 0 0.32 0.06 0.86 0.83 1 0.18 0.15 0.10 0.14 2 0.11 0.19 0.02 0.02 3 0.10 0.18 0.004 0.002 4 0.052 0.15 0.002 0.0002 5 0.032 0.10 0 2.4e-5 more actual zeros than predicted. For ERV, there are somewhat more actual zeros than fitted, but the difference is not too important. Why might OBDV not fit the zeros well? What if people made the decision to contact the doctor for a first visit, they are sick, then the doctor decides on whether or not follow-up visits are needed. This is a principal/agent type situation, where the total number of visits depends upon the decision of both the patient and the doctor. Since 464

25.1. HURDLE MODELS

465

different parameters may govern the two decision-makers choices, we might expect that different parameters govern the probability of zeros versus the other counts. Let λ p be the parameters of the patient’s demand for visits, and let λd be the paramter of the doctor’s “demand” for visits. The patient will initiate visits according to a discrete choice model, for example, a logit model:

Pr(Y = 0) = fY (0, λ p ) = 1 − 1/ [1 + exp(−λ p )] Pr(Y > 0)

1/ [1 + exp(−λ p )] ,

=

The above probabilities are used to estimate the binary 0/1 hurdle process. Then, for the observations where visits are positive, a truncated Poisson density is estimated. This density is fY (y, λd |y > 0) = =

fY (y, λd ) Pr(y > 0) fY (y, λd ) 1 − exp(−λd )

since according to the Poisson model with the doctor’s paramaters, Pr(y = 0) =

exp(−λd )λ0d . 0!

Since the hurdle and truncated components of the overall density for Y share no parameters, they may be estimated separately, which is computationally more efficient than estimating the overall model. (Recall that the BFGS algorithm, for example, will have to invert the approximated Hessian. The computational overhead is of order K 2

25.1. HURDLE MODELS

where K is the number of parameters to be estimated) . The expectation of Y is E(Y |x) = Pr(Y > 0|x)E(Y |Y > 0, x) 1 λd = 1 + exp(−λ p ) 1 − exp(−λd )

466

25.1. HURDLE MODELS

467

Here are hurdle Poisson estimation results for OBDV, obtained from this estimation program ************************************************************************** MEPS data, OBDV logit results Strong convergence Observations = 500 Function value

-0.58939

t-Stats params

t(OPG)

t(Sand.)

t(Hess)

constant

-1.5502

-2.5709

-2.5269

-2.5560

pub_ins

1.0519

3.0520

3.0027

3.0384

priv_ins

0.45867

1.7289

1.6924

1.7166

sex

0.63570

3.0873

3.1677

3.1366

age

0.018614

2.1547

2.1969

2.1807

educ

0.039606

1.0467

0.98710

1.0222

inc

0.077446

1.7655

2.1672

1.9601

Information Criteria Consistent Akaike 639.89 Schwartz 632.89 Hannan-Quinn 614.96 Akaike 603.39 **************************************************************************

25.1. HURDLE MODELS

468

The results for the truncated part: ************************************************************************** MEPS data, OBDV tpoisson results Strong convergence Observations = 500 Function value

-2.7042

t-Stats params

t(OPG)

t(Sand.)

constant

0.54254

7.4291

1.1747

3.2323

pub_ins

0.31001

6.5708

1.7573

3.7183

priv_ins

0.014382

0.29433

0.10438

0.18112

sex

0.19075

10.293

1.1890

3.6942

age

0.016683

16.148

3.5262

7.9814

educ

0.016286

4.2144

0.56547

1.6353

-0.0079016

-2.3186

-0.35309

-0.96078

inc

t(Hess)

Information Criteria Consistent Akaike 2754.7 Schwartz 2747.7 Hannan-Quinn 2729.8 Akaike 2718.2 **************************************************************************

25.1. HURDLE MODELS

469

Fitted and actual probabilites (NB-II fits are provided as well) are: TABLE 2. Actual and Hurdle Poisson fitted frequencies Count OBDV ERV Count Actual Fitted HP Fitted NB-II Actual Fitted HP Fitted NB-II 0 0.32 0.32 0.34 0.86 0.86 0.86 1 0.18 0.035 0.16 0.10 0.10 0.10 2 0.11 0.071 0.11 0.02 0.02 0.02 3 0.10 0.10 0.08 0.004 0.006 0.006 4 0.052 0.11 0.06 0.002 0.002 0.002 5 0.032 0.10 0.05 0 0.0005 0.001 For the Hurdle Poisson models, the ERV fit is very accurate. The OBDV fit is not so good. Zeros are exact, but 1’s and 2’s are underestimated, and higher counts are overestimated. For the NB-II fits, performance is at least as good as the hurdle Poisson model, and one should recall that many fewer parameters are used. Hurdle version of the negative binomial model are also widely used. 25.1.1. Finite mixture models. The following are results for a mixture of 2 negative binomial (NB-I) models, for the OBDV data, which you can replicate using this estimation program

25.1. HURDLE MODELS

470

************************************************************************** MEPS data, OBDV mixnegbin results Strong convergence Observations = 500 Function value

-2.2312

t-Stats params

t(OPG)

t(Sand.)

0.64852

1.3851

1.3226

1.4358

pub_ins

-0.062139

-0.23188

-0.13802

-0.18729

priv_ins

0.093396

0.46948

0.33046

0.40854

sex

0.39785

2.6121

2.2148

2.4882

age

0.015969

2.5173

2.5475

2.7151

educ

-0.049175

-1.8013

-1.7061

-1.8036

inc

0.015880

0.58386

0.76782

0.73281

ln_alpha

0.69961

2.3456

2.0396

2.4029

constant

-3.6130

-1.6126

-1.7365

-1.8411

pub_ins

2.3456

1.7527

3.7677

2.6519

priv_ins

0.77431

0.73854

1.1366

0.97338

sex

0.34886

0.80035

0.74016

0.81892

age

0.021425

1.1354

1.3032

1.3387

educ

0.22461

2.0922

1.7826

2.1470

inc

0.019227

0.20453

0.40854

0.36313

2.8419

6.2497

6.8702

7.6182

0.85186

1.7096

1.4827

1.7883

constant

ln_alpha logit_inv_mix

Information Criteria

t(Hess)

25.1. HURDLE MODELS

471

Consistent Akaike 2353.8 Schwartz 2336.8 Hannan-Quinn 2293.3 Akaike 2265.2 ************************************************************************** Delta method for mix parameter st. mix

se_mix

0.70096

0.12043

err.

• The 95% confidence interval for the mix parameter is perilously close to 1, which suggests that there may really be only one component density, rather than a mixture. Again, this is not the way to test this - it is merely suggestive. • Education is interesting. For the subpopulation that is “healthy”, i.e., that makes relatively few visits, education seems to have a positive effect on visits. For the “unhealthy” group, education has a negative effect on visits. The other results are more mixed. A larger sample could help clarify things. The following are results for a 2 component constrained mixture negative binomial model where all the slope parameters in λ j = exβ j are the same across the two components. The constants and the overdispersion parameters α j are allowed to differ for the two components.

25.1. HURDLE MODELS

472

************************************************************************** MEPS data, OBDV cmixnegbin results Strong convergence Observations = 500 Function value

-2.2441

t-Stats params

t(OPG)

t(Sand.)

constant

-0.34153

-0.94203

-0.91456

-0.97943

pub_ins

0.45320

2.6206

2.5088

2.7067

priv_ins

0.20663

1.4258

1.3105

1.3895

sex

0.37714

3.1948

3.4929

3.5319

age

0.015822

3.1212

3.7806

3.7042

educ

0.011784

0.65887

0.50362

0.58331

inc

0.014088

0.69088

0.96831

0.83408

ln_alpha

1.1798

4.6140

7.2462

6.4293

const_2

1.2621

0.47525

2.5219

1.5060

lnalpha_2

2.7769

1.5539

6.4918

4.2243

logit_inv_mix

2.4888

0.60073

3.7224

1.9693

Information Criteria Consistent Akaike 2323.5 Schwartz 2312.5 Hannan-Quinn

t(Hess)

25.2. MODELS FOR TIME SERIES DATA

473

2284.3 Akaike 2266.1 ************************************************************************** Delta method for mix parameter st. mix

se_mix

0.92335

0.047318

err.

• Now the mixture parameter is even closer to 1. • The slope parameter estimates are pretty close to what we got with the NB-I model. 25.2. Models for time series data This section can be ignored in its present form. Just left in to form a basis for completion (by someone else ?!) at some point. Hamilton, Time Series Analysis is a good reference for this section. This is very incomplete and contributions would be very welcome. Up to now we’ve considered the behavior of the dependent variable yt as a function of other variables xt . These variables can of course contain lagged dependent variables, e.g., xt = (wt , yt−1 , ..., yt− j ). Pure time series methods consider the behavior of yt as a function only of its own lagged values, unconditional on other observable variables. One can think of this as modeling the behavior of yt after marginalizing out all other variables. While it’s not immediately clear why a model that has other explanatory variables should marginalize to a linear in the parameters time series model, most time series work is done with linear models, though nonlinear time series is also a large and growing field. We’ll stick with linear time series models. 25.2.1. Basic concepts.


474

D EFINITION 53 (Stochastic process). A stochastic process is a sequence of random variables, indexed by time: ∞ {Yt }t=−∞

(25.2.1)

D EFINITION 54 (Time series). A time series is one observation of a stochastic process, over a specific interval: n {yt }t=1

(25.2.2)

So a time series is a sample of size n from a stochastic process. It’s important to keep in mind that conceptually, one could draw another sample, and that the values would be different.

D EFINITION 55 (Autocovariance). The jth autocovariance of a stochastic process is (25.2.3)

γ jt = E (yt − µt )(yt− j − µt− j )

where µt = E (yt ) .

D EFINITION 56 (Covariance (weak) stationarity). A stochastic process is covariance stationary if it has time constant mean and autocovariances of all orders: µt

= µ, ∀t

γ jt = γ j , ∀t As we’ve seen, this implies that γ j = γ− j : the autocovariances depend only one the interval between observations, but not the time of the observations.


475

D EFINITION 57 (Strong stationarity). A stochastic process is strongly stationary if the joint distribution of an arbitrary collection of the {Yt } doesn’t depend on t. Since moments are determined by the distribution, strong stationarity⇒weak stationarity. What is the mean of Yt ? The time series is one sample from the stochastic process. One could think of M repeated samples from the stoch. proc., e.g., {ytm } By a LLN, we would expect that

1 M p ytm → E (Yt ) ∑ M→∞ M m=1 lim

The problem is, we have only one sample to work with, since we can’t go back in time and collect another. How can E (Yt ) be estimated then? It turns out that ergodicity is the needed property. D EFINITION 58 (Ergodicity). A stationary stochastic process is ergodic (for the mean) if the time average converges to the mean (25.2.4)

1 n p ∑ yt → µ n t=1

A sufficient condition for ergodicity is that the autocovariances be absolutely summable:

∞

∑ |γ j | < ∞

j=0

This implies that the autocovariances die off, so that the yt are not so strongly dependent that they don’t satisfy a LLN. D EFINITION 59 (Autocorrelation). The jth autocorrelation, ρ j is just the jth autocovariance divided by the variance: (25.2.5)

ρj =

γj γ0


476

D EFINITION 60 (White noise). White noise is just the time series literature term for a classical error. εt is white noise if i) E (εt ) = 0, ∀t, ii) V (εt ) = σ2 , ∀t, and iii) εt

and εs are independent, t 6= s. Gaussian white noise just adds a normality assumption.

25.2.2. ARMA models. With these concepts, we can discuss ARMA models. These are closely related to the AR and MA error processes that we’ve already discussed. The main difference is that the lhs variable is observed directly now. MA(q) processes. A qth order moving average (MA) process is yt = µ + εt + θ1 εt−1 + θ2 εt−2 + · · · + θq εt−q where εt is white noise. The variance is = E (yt − µ)2

γ0

= E εt + θ1 εt−1 + θ2 εt−2 + · · · + θq εt−q = σ2 1 + θ21 + θ22 + · · · + θ2q Similarly, the autocovariances are

2

γ j = θ j + θ j+1 θ1 + θ j+2 θ2 + · · · + θq θq− j , j ≤ q = 0, j > q Therefore an MA(q) process is necessarily covariance stationary and ergodic, as long as σ2 and all of the θ j are finite. AR(p) processes. An AR(p) process can be represented as yt = c + φ1 yt−1 + φ2 yt−2 + · · · + φ p yt−p + εt


477

The dynamic behavior of an AR(p) process can be studied by writing this pth order difference equation as a vector first order difference equation:       φ φ ··· φp   1 2    yt−1  yt   c   1 0 0 0        yt−1   0     yt−2     .  .. 0  .  =  .  0 1 0   ..  ..   ..    .      .. . . . .. .. .. 0···      .   yt−p  yt−p+1 0 0 ··· 0 1 0





        +      



εt   0   ..  .    0

or

Yt = C + FYt−1 + Et With this, we can recursively work forward in time: Yt+1

= C + FYt + Et+1 = C + F (C + FYt−1 + Et ) + Et+1 = C + FC + F 2Yt−1 + FEt + Et+1

and Yt+2

= C + FYt+1 + Et+2 = C + F C + FC + F 2Yt−1 + FEt + Et+1 + Et+2

= C + FC + F 2C + F 3Yt−1 + F 2 Et + FEt+1 + Et+2 or in general Yt+ j = C + FC + · · · + F jC + F j+1Yt−1 + F j Et + F j−1 Et+1 + · · · + FEt+ j−1 + Et+ j


478

Consider the impact of a shock in period t on yt+ j . This is simply ∂Yt+ j j = F(1,1) 0 ∂Et (1,1) If the system is to be stationary, then as we move forward in time this impact must die off. Otherwise a shock causes a permanent change in the mean of yt . Therefore, stationarity requires that j lim F j→∞ (1,1)

=0

• Save this result, we’ll need it in a minute. Consider the eigenvalues of the matrix F. These are the for λ such that |F − λIP | = 0 The determinant here can be expressed as a polynomial. for example, for p = 1, the matrix F is simply F = φ1 so |φ1 − λ| = 0 can be written as φ1 − λ = 0 When p = 2, the matrix F is

so



φ1 φ2



φ1 − λ φ 2

F=

F − λIP = 

1

1

0

 

−λ

 


479

and |F − λIP | = λ2 − λφ1 − φ2 So the eigenvalues are the roots of the polynomial λ2 − λφ1 − φ2 which can be found using the quadratic equation. This generalizes. For a pth order AR process, the eigenvalues are the roots of λ p − λ p−1 φ1 − λ p−2 φ2 − · · · − λφ p−1 − φ p = 0 Supposing that all of the roots of this polynomial are distinct, then the matrix F can be factored as F = T ΛT −1 where T is the matrix which has as its columns the eigenvectors of F, and Λ is a diagonal matrix with the eigenvalues on the main diagonal. Using this decomposition, we can write F j = T ΛT −1

T ΛT −1 · · · T ΛT −1

where T ΛT −1 is repeated j times. This gives

F j = T Λ j T −1 and



j λ1

   0  j Λ =    0

0 j

λ2



0      ..  .   j λp


480

Supposing that the λi i = 1, 2, ..., p are all real valued, it is clear that j

lim F(1,1) = 0

j→∞

requires that |λi | < 1, i = 1, 2, ..., p e.g., the eigenvalues must be less than one in absolute value.

• It may be the case that some eigenvalues are complex-valued. The previous result generalizes to the requirement that the eigenvalues be less than one in modulus, where the modulus of a complex number a + bi is mod(a + bi) =

p a2 + b2

This leads to the famous statement that “stationarity requires the roots of the determinantal polynomial to lie inside the complex unit circle.” draw picture here. • When there are roots on the unit circle (unit roots) or outside the unit circle, we leave the world of stationary processes. j

• Dynamic multipliers: ∂yt+ j /∂εt = F(1,1) is a dynamic multiplier or an impulseresponse function. Real eigenvalues lead to steady movements, whereas comlpex eigenvalue lead to ocillatory behavior. Of course, when there are multiple eigenvalues the overall effect can be a mixture. pictures

Invertibility of AR process To begin with, define the lag operator L Lyt = yt−1


481

The lag operator is defined to behave just as an algebraic quantity, e.g., L2 yt = L(Lyt ) = Lyt−1 = yt−2 or (1 − L)(1 + L)yt = 1 − Lyt + Lyt − L2 yt = 1 − yt−2 A mean-zero AR(p) process can be written as yt − φ1 yt−1 − φ2 yt−2 − · · · − φ p yt−p = εt or yt (1 − φ1 L − φ2 L2 − · · · − φ p L p ) = εt Factor this polynomial as 1 − φ1 L − φ2 L2 − · · · − φ p L p = (1 − λ1 L)(1 − λ2 L) · · · (1 − λ p L) For the moment, just assume that the λi are coefficients to be determined. Since L is defined to operate as an algebraic quantitiy, determination of the λ i is the same as determination of the λi such that the following two expressions are the same for all z : 1 − φ1 z − φ2 z2 − · · · − φ p z p = (1 − λ1 z)(1 − λ2 z) · · ·(1 − λ p z)


482

Multiply both sides by z−p z−p − φ1 z1−p − φ2 z2−p − · · · φ p−1 z−1 − φ p = (z−1 − λ1 )(z−1 − λ2 ) · · · (z−1 − λ p ) and now define λ = z−1 so we get λ p − φ1 λ p−1 − φ2 λ p−2 − · · · − φ p−1 λ − φ p = (λ − λ1 )(λ − λ2 ) · · · (λ − λ p ) The LHS is precisely the determinantal polynomial that gives the eigenvalues of F. Therefore, the λi that are the coefficients of the factorization are simply the eigenvalues of the matrix F. Now consider a different stationary process (1 − φL)yt = εt • Stationarity, as above, implies that |φ| < 1. Multiply both sides by 1 + φL + φ2 L2 + ... + φ j L j to get 1 + φL + φ2 L2 + ... + φ j L j (1 − φL)yt = 1 + φL + φ2 L2 + ... + φ j L j εt

or, multiplying the polynomials on th LHS, we get

1 + φL + φ2 L2 + ... + φ j L j − φL − φ2 L2 − ... − φ j L j − φ j+1 L j+1 yt == 1 + φL + φ2 L2 + ... + φ j L j εt

and with cancellations we have

so

1 − φ j+1 L j+1 yt = 1 + φL + φ2 L2 + ... + φ j L j εt yt = φ j+1 L j+1 yt + 1 + φL + φ2 L2 + ... + φ j L j εt


483

Now as j → ∞, φ j+1 L j+1 yt → 0, since |φ| < 1, so yt ∼ = 1 + φL + φ2 L2 + ... + φ j L j εt

and the approximation becomes better and better as j increases. However, we started with (1 − φL)yt = εt Substituting this into the above equation we have

so

yt ∼ = 1 + φL + φ2 L2 + ... + φ j L j (1 − φL)yt 1 + φL + φ2 L2 + ... + φ j L j (1 − φL) ∼ =1

and the approximation becomes arbitrarily good as j increases arbitrarily. Therefore, for |φ| < 1, define

(1 − φL)−1 =

∞

∑ φ jL j

j=0

Recall that our mean zero AR(p) process yt (1 − φ1 L − φ2 L2 − · · · − φ p L p ) = εt can be written using the factorization yt (1 − λ1 L)(1 − λ2 L) · · · (1 − λ p L) = εt where the λ are the eigenvalues of F, and given stationarity, all the |λi | < 1. Therefore, we can invert each first order polynomial on the LHS to get ! ! yt =

∞

∑ λ1 L j

j=0

j

∞

∑ λ2 L j

j=0

j

∞

···

∑ λ pj L j

j=0

!

εt


484

The RHS is a product of infinite-order polynomials in L, which can be represented as yt = (1 + ψ1 L + ψ2 L2 + · · · )εt where the ψi are real-valued and absolutely summable. • The ψi are formed of products of powers of the λi , which are in turn functions of the φi .

• The ψi are real-valued because any complex-valued λi always occur in conjugate pairs. This means that if a + bi is an eigenvalue of F, then so is a − bi. In multiplication (a + bi) (a − bi) = a2 − abi + abi − b2 i2 = a2 + b2 which is real-valued. • This shows that an AR(p) process is representable as an infinite-order MA(q) process. • Recall before that by recursive substitution, an AR(p) process can be written as Yt+ j = C + FC + · · · + F jC + F j+1Yt−1 + F j Et + F j−1 Et+1 + · · · + FEt+ j−1 + Et+ j If the process is mean zero, then everything with a C drops out. Take this and lag it by j periods to get Yt = F j+1Yt− j−1 + F j Et− j + F j−1 Et− j+1 + · · · + FEt−1 + Et As j → ∞, the lagged Y on the RHS drops out. The Et−s are vectors of zeros except for their first element, so we see that the first equation here, in the


limit, is just yt =

∞

∑

j=0

Fj

485

ε 1,1 t− j

which makes explicit the relationship between the ψi and the φi (and the λi as well, recalling the previous factorization of F j ).

Moments of AR(p) process. The AR(p) process is yt = c + φ1 yt−1 + φ2 yt−2 + · · · + φ p yt−p + εt Assuming stationarity, E (yt ) = µ, ∀t, so µ = c + φ1 µ + φ2 µ + ... + φ p µ so µ= and

c 1 − φ1 − φ2 − ... − φ p

c = µ − φ1 µ − ... − φ p µ so yt − µ = µ − φ1 µ − ... − φ p µ + φ1 yt−1 + φ2 yt−2 + · · · + φ p yt−p + εt − µ = φ1 (yt−1 − µ) + φ2 (yt−2 − µ) + ... + φ p (yt−p − µ) + εt With this, the second moments are easy to find: The variance is γ0 = φ1 γ1 + φ2 γ2 + ... + φ p γ p + σ2


486

The autocovariances of orders j ≥ 1 follow the rule γj

= E (yt − µ) yt− j − µ)

= E (φ1 (yt−1 − µ) + φ2 (yt−2 − µ) + ... + φ p (yt−p − µ) + εt ) yt− j − µ = φ1 γ j−1 + φ2 γ j−2 + ... + φ p γ j−p

Using the fact that γ− j = γ j , one can take the p + 1 equations for j = 0, 1, ..., p, which have p + 1 unknowns (σ2 , γ0 , γ1 , ..., γ p ) and solve for the unknowns. With these, the γ j for j > p can be solved for recursively. Invertibility of MA(q) process. An MA(q) can be written as yt − µ = (1 + θ1 L + ... + θq Lq )εt As before, the polynomial on the RHS can be factored as (1 + θ1 L + ... + θq Lq ) = (1 − η1 L)(1 − η2 L)...(1 − ηq L) and each of the (1 − ηi L) can be inverted as long as |ηi | < 1. If this is the case, then we can write (1 + θ1 L + ... + θq Lq )−1 (yt − µ) = εt where (1 + θ1 L + ... + θq Lq )−1 will be an infinite-order polynomial in L, so we get ∞

∑ −δ j L j (yt− j − µ) = εt

j=0


487

with δ0 = −1, or (yt − µ) − δ1 (yt−1 − µ) − δ2 (yt−2 − µ) + ... = εt or yt = c + δ1 yt−1 + δ2 yt−2 + ... + εt where c = µ + δ1 µ + δ2 µ + ... So we see that an MA(q) has an infinite AR representation, as long as the |η i | < 1, i = 1, 2, ..., q.

• It turns out that one can always manipulate the parameters of an MA(q) process to find an invertible representation. For example, the two MA(1) processes yt − µ = (1 − θL)εt and yt∗ − µ = (1 − θ−1 L)εt∗ have exactly the same moments if σ2ε∗ = σ2ε θ2 For example, we’ve seen that γ0 = σ2 (1 + θ2 ). Given the above relationships amongst the parameters, γ∗0 = σ2ε θ2 (1 + θ−2 ) = σ2 (1 + θ2 )


488

so the variances are the same. It turns out that all the autocovariances will be the same, as is easily checked. This means that the two MA processes are observationally equivalent. As before, it’s impossible to distinguish between observationally equivalent processes on the basis of data. • For a given MA(q) process, it’s always possible to manipulate the parameters to find an invertible representation (which is unique). • It’s important to find an invertible representation, since it’s the only repre-

sentation that allows one to represent εt as a function of past y0 s. The other representations express

• Why is invertibility important? The most important reason is that it provides a justification for the use of parsimonious models. Since an AR(1) process has an MA(∞) representation, one can reverse the argument and note that at least some MA(∞) processes have an AR(1) representation. At the time of estimation, it’s a lot easier to estimate the single AR(1) coefficient rather than the infinite number of coefficients associated with the MA representation. • This is the reason that ARMA models are popular. Combining low-order AR and MA models can usually offer a satisfactory representation of univariate time series data with a reasonable number of parameters. • Stationarity and invertibility of ARMA models is similar to what we’ve seen - we won’t go into the details. Likewise, calculating moments is similar. E XERCISE 61. Calculate the autocovariances of an ARMA(1,1) model: (1+φL)yt = c + (1 + θL)εt

Bibliography [1] Davidson, R. and J.G. MacKinnon (1993) Estimation and Inference in Econometrics, Oxford Univ. Press. [2] Davidson, R. and J.G. MacKinnon (2004) Econometric Theory and Methods, Oxford Univ. Press. [3] Gallant, A.R. (1985) Nonlinear Statistical Models, Wiley. [4] Gallant, A.R. (1997) An Introduction to Econometric Theory, Princeton Univ. Press. [5] Hamilton, J. (1994) Time Series Analysis, Princeton Univ. Press [6] Hayashi, F. (2000) Econometrics, Princeton Univ. Press. [7] Wooldridge (2003), Introductory Econometrics, Thomson. (undergraduate level, for supplementary use only).

489

Index

asymptotic equality, 452

observations, influential, 33 outliers, 33 own influence, 34

Chain rule, 447 Cobb-Douglas model, 27

parameter space, 54

convergence, almost sure, 449

Product rule, 447

convergence, in distribution, 449 convergence, in probability, 449

R- squared, uncentered, 36

Convergence, ordinary, 448

R-squared, centered, 37

convergence, pointwise, 448 convergence, uniform, 448 convergence, uniform almost sure, 450 cross section, 23

estimator, linear, 33, 43 estimator, OLS, 29 extremum estimator, 242

leverage, 34 likelihood function, 54

matrix, idempotent, 33 matrix, projection, 32 matrix, symmetric, 33 490

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics,

Econometrics

Econometrics

Econometrics

Bayesian Econometrics (Advances in Econometrics)

Introductory Econometrics

Spatial Econometrics

Basic Econometrics

Financial Econometrics

Advanced Econometrics

Advanced Econometrics

Ebooks Econometrics The Econometrics of Financial Markets

Finite Sample Econometrics (Advanced Texts in Econometrics)

Bayesian econometrics

Advanced Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics

Econometrics,

Econometrics

Econometrics

Econometrics

Bayesian Econometrics (Advances in Econometrics)

Introductory Econometrics

Spatial Econometrics

Basic Econometrics

Financial Econometrics

Advanced Econometrics

Advanced Econometrics

Ebooks Econometrics The Econometrics of Financial Markets

Finite Sample Econometrics (Advanced Texts in Econometrics)

Bayesian econometrics

Advanced Econometrics

Recommend Documents